bandwidth study on energy use and potential energy … study on energy use and potential energy...

Bandwidth Study on Energy Use and Potential Energy Saving Opportunities in U.S. Petroleum Refining i

,

ii Bandwidth Study on Energy Use and Potential Energy Saving Opportunities in U.S. Petroleum Refining

THE U.S. DEPARTMENT OF ENERGY (DOE)’S ADVANCED MANUFACTURING OFFICE

PROVIDED FUNDING FOR THIS ANALYSIS AND REPORT

The DOE Office of Energy Efficiency and Renewable Energy (EERE)’s Advanced

Manufacturing Office works with industry, small business, universities, and other stakeholders to

identify and invest in emerging technologies with the potential to create high-quality domestic

manufacturing jobs and enhance the global competitiveness of the United States.

Prepared for DOE / EERE’s Advanced Manufacturing Office by Energetics Incorporated

(Sabine Brueske, Caroline Kramer, Aaron Fisher)

Disclaimer

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United

States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any

legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process

disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial

product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its

endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of

authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

Bandwidth Study on Energy Use and Potential Energy Saving Opportunities in U.S. Petroleum Refining iii

Preface

Reducing energy consumption through investment in advanced technologies and practices can

enhance American manufacturing competitiveness. Energy bandwidth studies of U.S.

manufacturing sectors serve as general data references to help understand the range (or

bandwidth) of potential energy savings opportunities. The U.S. Department of Energy (DOE)’s

Advanced Manufacturing Office (AMO) has commissioned a series of bandwidth studies to

analyze the processes and products that consume the most energy, and provide hypothetical,

technology-based estimates of potential energy savings opportunities. The consistent

methodology used in the bandwidth studies provides a framework to evaluate and compare

energy savings potentials within and across manufacturing sectors at the macro-scale. Bandwidth

studies using the terminology and methodology outlined below were prepared for the Chemicals,

Petroleum Refining, Iron and Steel, and Pulp and Paper industry sectors in 2014.11

Four different energy bands (or

measures) are used consistently in this

series to describe different levels of

onsite energy consumption to

manufacture specific products and to

compare potential energy savings

opportunities in U.S. manufacturing

facilities (see figure). Current typical

(CT) is the energy consumption in

2010; state of the art (SOA) is the

energy consumption that may be

possible through the adoption of

existing best technologies and practices

available worldwide; practical

minimum (PM) is the energy

consumption that may be possible if

applied R&D technologies under

development worldwide are deployed;

and the thermodynamic minimum

(TM) is the least amount of energy required under ideal conditions, which typically cannot be

attained in commercial applications. CT energy consumption serves as the benchmark of

manufacturing energy consumption. TM energy consumption serves as the baseline (or

1 The concept of an energy bandwidth, and its use as an analysis tool for identifying potential energy saving opportunities,

originated in AMO in 2002 (when it was called the Office of Industrial Technologies). The first two sector studies—Iron and

Steel, and Metal Castings—were completed in 2004. That work was followed by Chemicals and Petroleum Refining studies in

2006, and Aluminum, Glass, and Mining in 2007. A Cement Industry analysis was conducted in 2010 and a Pulp and Paper

analysis was conducted in 2011.

Energy Consumption Bands and Opportunity

Bandwidths Estimated in this Study

iv Bandwidth Study on Energy Use and Potential Energy Saving Opportunities in U.S. Petroleum Refining

theoretical minimum) that is used in calculating energy savings potential. Feedstock energy (the

nonfuel use of fossil energy) is not included in the energy consumption estimates.

Two onsite energy savings opportunity bandwidths are estimated: the current opportunity spans

the bandwidth from CT energy consumption to SOA energy consumption, and the R&D

opportunity spans the bandwidth from SOA energy consumption to PM energy consumption.

These bandwidths are estimated for processes and products studied and for all manufacturing

within a sector based on extrapolated data. The difference between PM energy consumption and

TM energy consumption is labeled as impractical. The term impractical is used because with

today’s knowledge of technologies in R&D, further investment may no longer lead to

incremental energy savings and thermodynamic limitations impede technology opportunities.

Significant investment in technology development and implementation would be needed to fully

realize the energy savings opportunities estimated. The costs associated with achieving SOA and

PM energy consumption are not considered in this report; a techno-economic analysis of the

costs and benefits of R&D technologies was not in the scope of this study.

In each sector studied in the series, the four energy bands are estimated for select individual

products or processes, subsectors, and sector-wide. The estimation method compares diverse

industry, governmental, and academic data to analyses of reported plant energy consumption

data from the Manufacturing Energy Consumption Survey (MECS) conducted by the U.S.

Energy Information Administration (EIA). MECS is a national sample survey of U.S.

manufacturing establishments conducted every four years; information is collected and reported

on U.S. manufacturing energy consumption and expenditures.

Acknowledgements

AMO wishes to acknowledge the contributions made by Jimmy Kumana of Kumana and

Associates for his work reviewing this study. Appreciated is also extended to UOP, a Honeywell

Company, for their assistance with estimating research and development savings.

In addition, AMO recognizes Joseph Cresko of DOE/AMO who lead the conceptual

development and publication of the bandwidth study series with the support of Dr. Alberta

Carpenter at the National Renewable Energy Laboratory, as well as the important contributions

made by Sabine Brueske, Caroline Kramer, and Dr. Aaron Fisher of Energetics Incorporated for

conducting the majority of the research and analysis and drafting this study.

Bandwidth Study on Energy Use and Potential Energy Saving Opportunities in U.S. Petroleum Refining v

Executive Summary

The United States was the largest producer of refined petroleum products in 2010, with a net

production of a total of over 12.5 million barrels per day (EIA 2013b). This bandwidth study

examines energy consumption and potential energy savings opportunities in U.S. petroleum

refining. Industrial, government, and academic data are used to estimate the energy consumed in

nine of the most energy intensive petroleum refining processes. Three different energy

consumption bands (or levels) are estimated for these select manufacturing processes based on

referenced energy intensities of current, state of the art, and R&D technologies. A fourth

theoretical minimum energy consumption band is also estimated. The data from the select

processes studied is also extrapolated to determine energy consumption for the entire petroleum

refining sector. The bandwidth—the difference between bands of energy consumption—is used

to determine the potential energy savings opportunity. The costs associated with realizing these

energy savings was not in the scope of this study.

The purpose of this data analysis is to provide macro-scale estimates of energy savings

opportunities for petroleum refining processes and sector-wide. This is a step toward

understanding the processes that could most benefit from technology and efficiency

improvements to realize energy savings.

Study Organization and Approach: After providing an overview of the methodology (Chapter 1)

and energy consumption in petroleum refining (Chapter 2), the 2010 throughput and production

volumes (Chapter 3) and current energy consumption (current typical [CT], Chapter 4) were

estimated for nine select processes. In addition, the minimum energy consumption for these

processes was estimated assuming the adoption of best technologies and practices available

worldwide (state of the art [SOA], Chapter 5) and assuming the deployment of the applied

research and development (R&D) technologies available worldwide (practical minimum [PM],

Chapter 6). The minimum amount of energy theoretically required for these processes assuming

ideal conditions was also estimated (thermodynamic minimum [TM)], Chapter 7); in some cases,

this is less than zero. The difference between the energy consumption bands (CT, SOA, PM,

TM) are the estimated energy savings opportunity bandwidths (Chapter 8).

The U.S. Energy Information Administration’s (EIA) Manufacturing Energy Consumption

Survey (MECS) provides a sector-wide estimate of energy consumption for U.S. petroleum

refining; this data is referenced as sector-wide CT energy consumption. In this study, CT, SOA,

PM, and TM energy consumption for nine individual processes is estimated from multiple

referenced sources. To estimate SOA, PM, and TM energy consumption for the entire sector, the

CT, SOA, PM, and TM energy consumption data of the nine processes studied is extrapolated

estimate total sector-wide SOA, PM, and TM energy consumption. In 2010, these nine processes

corresponded to 68% of the industry’s energy consumption.

vi Bandwidth Study on Energy Use and Potential Energy Saving Opportunities in U.S. Petroleum Refining

Study Results: Two energy savings opportunity bandwidths – current opportunity and R&D

opportunity – are presented in Table ES-1 and Figure ES-1.1 The current opportunity is the

difference between the 2010 CT energy consumption and SOA energy consumption; the R&D

opportunity is the difference between SOA energy consumption and PM energy consumption.

Potential energy savings opportunities are presented for the nine processes studied and for all of

U.S. petroleum refining based on extrapolated data. Figure ES-1 also shows the estimated

relative current and R&D energy savings opportunities for individual processes based on the

sector-wide extrapolated data.

Table ES-1. Potential Energy Savings Opportunities in the U.S. Petroleum Refining Sector[1]

Opportunity Bandwidths

Estimated Energy Savings

Opportunity for Nine Select Petroleum

Refining Processes

(per year)

Estimated Energy Savings

Opportunity for

All of the

U.S. Petroleum Refining Sector

Based on

Extrapolated Data

(per year)

Current Opportunity – energy

savings if the best technologies

and practices available are used to

upgrade production

286 TBtu2

(14% energy savings,

where TM is the baseline)

420 TBtu3



R&D Opportunity – additional

energy savings if the applied R&D

technologies under development

worldwide are deployed

540 TBtu4



793 TBtu5



1 The energy estimates presented in this study are for macro-scale consideration; energy intensities and energy

consumption values do not represent energy use in any specific facility or any particular region in the United States.

The costs associated with achieving energy savings are not considered in this study. All estimates are for onsite

energy use (i.e., energy consumed within the refinery boundary). Energy used as feedstocks (non-fuel inputs) to

production is excluded. 2 286 TBtu = 2163 – 1877

3 420 TBtu = 3176 – 2756

4 540 TBtu = 1877 – 1336

5 793 TBtu = 2756 – 1963

Bandwidth Study on Energy Use and Potential Energy Saving Opportunities in U.S. Petroleum Refining vii

The PM energy consumption estimates are speculative because they are based on unproven

technologies. The estimates assume deployment of R&D technologies that are under

development; where multiple technologies were considered for a similar application, only the

most energy efficient technology was considered in the energy savings estimate. The difference

between PM and TM is labeled “impractical” because with today’s knowledge of technologies in

R&D, further investment may no longer lead to incremental energy savings and thermodynamic

limitations impede technology opportunities.

The results presented show that 286 TBtu of energy could be saved each year if capital

investments in the best technologies and practices available worldwide are used to upgrade nine

petroleum refining processes; an additional 540 TBtu could be saved through the adoption of

applied R&D technologies under development worldwide.

However, if the energy savings potential is estimated for the U.S. petroleum refining industry as

a whole, the current energy savings opportunity is 420 TBtu per year and the R&D opportunity

increases to 793 TBtu per year.

Figure ES-1. Current and R&D Energy Savings Opportunities for the Nine Processes Studied and for Petroleum

Refining Sector-wide Based on Extrapolated Data

viii Bandwidth Study on Energy Use and Potential Energy Saving Opportunities in U.S. Petroleum Refining

The top four Current Energy Savings Opportunities for the processes are as follows:

All other NAICS 3241101 processes - 134 TBtu (or 32% of the current opportunity)

Atmospheric Crude Distillation - 82 TBtu (or 20% of the current opportunity)

Hydrotreating - 57 TBtu (or 14% of the current opportunity) and

Fluid Catalytic Cracking - 45 TBtu (or 11% of the current opportunity).

The top four R&D Energy Saving Opportunities for the processes are as follows:

All other NAICS 3241102 processes - 253 TBtu (or 32% of the R&D opportunity)

Atmospheric Crude Distillation - 208 TBtu (or 26% of the R&D opportunity)

Hydrotreating - 81 TBtu (or 10% of the R&D opportunity) and

Catalytic Reforming - 58 TBtu (or 7% of the R&D opportunity).

1 All other NAICS 324110 includes all other processes in the petroleum refining sector other than the nine processes

studied. 2 All other NAICS 324110 includes all other processes in the petroleum refining sector other than the nine processes

studied.

Bandwidth Study on Energy Use and Potential Energy Saving Opportunities in U.S. Petroleum Refining ix

List of Acronyms and Abbreviations

AMO Advanced Manufacturing Office

bbl Barrel

BPSD Barrels per stream day

Btu British thermal unit

C Carbon

Cp Specific heat

CDU Crude (oil) distillation unit

CHP Combined heat and power

CT Current typical energy consumption or energy intensity

DOE U.S. Department of Energy

EIA U.S. Energy Information Administration

EERE DOE Office of Energy Efficiency and Renewable Energy

FCC Fluid catalytic cracking

G Gibbs free energy

GEMS Global Energy Management System

H Enthalpy or hydrogen

HEN Heat exchanger network

HVAC Heating, ventilation, and air conditioning

K degrees Kelvin (as suffix to numerical temperature measure)

LPG Liquefied petroleum gases

MECS Manufacturing Energy Consumption Survey

NAICS North American Industry Classification System

NGL Natural gas liquids

PM Practical minimum energy consumption or energy intensity

ppm Parts per million

R&D Research and development

S Entropy

scf Standard cubic feet

SOA State of the art energy consumption or energy intensity

T Temperature

TBtu Trillion British thermal units

TM Thermodynamic minimum energy consumption or energy intensity

Wt% Weight percent

x Bandwidth Study on Energy Use and Potential Energy Saving Opportunities in U.S. Petroleum Refining

Table of Contents

Preface .......................................................................................................................... iii

Executive Summary ...................................................................................................... v

List of Acronyms and Abbreviations .......................................................................... ix

1. Introduction .............................................................................................................. 1

1.1. Overview .........................................................................................................................1

1.2. Comparison to Other Bandwidth Studies ........................................................................1

1.3. Definitions of Energy Consumption Bands and Opportunity Bandwidths .....................2

1.4. Bandwidth Analysis Method ...........................................................................................3

2. U.S. Petroleum Refining Sector Overview ............................................................. 7

2.1. U.S. Petroleum Refining Economic Overview ...............................................................7

2.2. U.S. Petroleum Refining Products, Establishments, and Processes ................................7

2.3. U.S. Petroleum Refining Energy Consumption ............................................................11

2.3.1. Fuel and Feedstocks ............................................................................................13

2.3.2. Electricity ............................................................................................................15

2.3.3. Steam...................................................................................................................16

3. Throughputs and Production in U.S. Petroleum Refining .................................. 17

4. Current Typical Energy Consumption for U.S. Petroleum Refining .................. 19

4.1. Boundaries of the Petroleum Refining Bandwidth Study .............................................19

4.2. Estimated Energy Intensity for Individual Processes ....................................................19

4.3. Calculated Current Typical Energy Consumption for Individual Processes .................21

4.4. Current Typical Energy Consumption By Process and Sector-wide ............................23

5. State of the Art Energy Consumption for U.S. Petroleum Refining ................... 25

5.1. Calculated State of the Art Energy Consumption for Individual Processes ..................25

5.2. State of the Art Energy Consumption By Process and Sector-wide .............................29

5.2.1. Comparing State of the Art and Current Typical Energy Data ...........................32

5.2.1.1. Atmospheric Crude Distillation ............................................................ 32

5.2.1.2. Fluid Catalytic Cracking ...................................................................... 33

5.2.1.3. Catalytic Reforming ............................................................................. 33

5.2.1.4. Vacuum Crude Distillation ................................................................... 33

6. Practical Minimum Energy Consumption for U.S. Petroleum Refining ............. 35

6.1. R&D in the Petroleum Refining Industry ......................................................................35

6.2. Description of A Future Refinery ..................................................................................36

6.3. Calculated Practical Minimum Energy Consumption for Individual Processes ...........37

6.3.1. Weighting of Technologies .................................................................................39

Bandwidth Study on Energy Use and Potential Energy Saving Opportunities in U.S. Petroleum Refining xi

6.4. Practical Minimum Energy Consumption By Process and Sector-wide .......................40

6.4.1. Atmospheric and Vacuum Crude Distillation Practical Minimum Energy

Savings ..........................................................................................................................45

7. Thermodynamic Minimum Energy Consumption for U.S. Petroleum

Refining .................................................................................................................. 47

7.1. Thermodynamic Minimum Energy ...............................................................................47

7.2. Calculated Thermodynamic Minimum Energy Consumption for Individual

Processes .......................................................................................................................47

7.2.1. Calculated Thermodynamic Minimum Energy Intensities .................................48

7.3. Thermodynamic Minimum Energy Consumption By Process and Sector-wide ..........49

8. U.S. Petroleum Refining Energy Bandwidth Summary ...................................... 51

8.1. Petroleum Refining Bandwidth Profile .........................................................................51

9. References ............................................................................................................. 55

Appendix A1: Master Petroleum Refining Table....................................................... 60

Appendix A2: References for U.S. Throughput/Production Data of the 9

Processes Studied and Energy Intensity Data Used to Calculate the

Current Typical, State of the Art, and Thermodynamic Minimum Energy

Consumption Bands .............................................................................................. 61

Appendix A3: Technologies Analyzed to Estimate Practical Minimum

Energy Intensities with References ...................................................................... 63

References for Table A3 .........................................................................................................69

Appendix A4: Practical Minimum Technology Weighting Factors ......................... 70

Methodology to Determine Weighting Factors ......................................................................70

Appendix A5: Thermodynamic Minimum Calculation Details ................................ 75

Atmospheric and Vacuum Crude Distillation Thermodynamic Minimum Energy

Intensity .........................................................................................................................75

Hydrotreating Thermodynamic Minimum Energy Intensity ..................................................78

Catalytic Hydrocracking Thermodynamic Minimum Energy Intensity .................................79

Isomerization Thermodynamic Minimum Energy Intensity ..................................................80

Coking/Visbreaking (Delayed Coking) Thermodynamic Minimum Energy Intensity ..........81

Thermodynamic Minimum Energy Intensity References ......................................................83

xii Bandwidth Study on Energy Use and Potential Energy Saving Opportunities in U.S. Petroleum Refining

List of Figures

Figure ES-1. Current and R&D Energy Savings Opportunities for the Nine Processes

Studied and for Petroleum Refining Sector-wide Based on Extrapolated

Data ....................................................................................................................... vii

Figure 1-1. Energy Consumption Bands and Opportunity Bandwidths Estimated in this

Study ....................................................................................................................... 2

Figure 2-1. Geographic Distribution of Petroleum Refineries in the United States

(created from EIA 2013d) ....................................................................................... 9

Figure 2-2. Typical Petroleum Refinery Flow Diagram (DOE 2007) .......................................... 10

Figure 2-3. Onsite Energy Entering U.S. Petroleum Refineries, 2010 (DOE 2014) .................... 12

Figure 2-4. Onsite Energy Consumption at Point of End Use in U.S. Petroleum

Refineries, 2010 (DOE 2014) ............................................................................... 13

Figure 2-5. Fuel Consumption in the Petroleum Refining Sector by End Use, 2010 (DOE

2014) ..................................................................................................................... 14

Figure 2-6. Electricity Consumption in the Petroleum Refining Sector by End Use, 2010

(DOE 2014) ........................................................................................................... 15

Figure 2-7. Steam Consumption in the Petroleum Refining Sector by End Use, 2010

(DOE 2014) ........................................................................................................... 16

Figure 5-1. Current Opportunity Energy Savings Bandwidths for Processes Studied (with

Percent of Overall Energy Consumption Bandwidth) .......................................... 31

Figure 5-2. Current Energy Savings Opportunity by Petroleum Refining Process Studied

(Energy Savings Per Year in TBtu) ....................................................................... 32

Figure 6-1. Current and R&D Opportunity Energy Savings Bandwidths for the Petroleum

Refining Processes Studied (with Percent of Overall Energy Consumption

Bandwidth) ............................................................................................................ 43

Figure 6-2. Current and R&D Energy Savings Opportunities by Petroleum Refining

Process Studied (Energy Savings Per Year in TBtu) ............................................ 44

Figure 8-1. Current and R&D Energy Savings Opportunities in U.S. Petroleum Refining

for the Processes Studied and for Sector-Wide based on Extrapolated Data ....... 53

Figure 8-2. Current and R&D Opportunity Energy Savings Bandwidths for the Processes

Studied (with Percent of Overall Energy Consumption Bandwidth) .................... 54

List of Tables

Table ES-1. Potential Energy Savings Opportunities in the U.S. Petroleum Refining

Sector[1]

.................................................................................................................. vi

Table 2-1. U.S. Petroleum Refinery Products, 2010....................................................................... 8

Table 2-2. Petroleum Refining Processes Selected for Bandwidth Analysis ............................... 10

Table 2-3. U.S. Petroleum Refining Energy Consumption Sector-Wide, 2010 ........................... 11

Table 3-1. U.S. Petroleum Refining Process Capacities and Throughputs/Production for

the Nine Processes Studied in 2010 ...................................................................... 17

Bandwidth Study on Energy Use and Potential Energy Saving Opportunities in U.S. Petroleum Refining xiii

Table 4-1. Published Sources Reviewed to Identify Current Typical Energy Intensities for

Processes Studied .................................................................................................. 21

Table 4-2. Calculated U.S. Onsite Current Typical Energy Consumption for Processes

Studied in 2010 with Calculated Primary Energy Consumption and Offsite

Losses .................................................................................................................... 22

Table 4-3. Onsite and Primary Current Typical Energy Consumption for the Nine

Processes Studied and Sector-Wide in 2010, with Percent of Sector

Coverage ............................................................................................................... 24

Table 5-1. Published Sources Referenced to Identify State of the Art Energy Intensities

for Nine Select Processes ...................................................................................... 26

Table 5-2. Consulted Petroleum Refining SOA Case Studies Breakdown and Details ............... 27

Table 5-3. Specific Petroleum Refining Process Unit Energy Efficiency Opportunities* ........... 28

Table 5-4. Onsite State of the Art Energy Consumption, Energy Savings, and Energy

Savings Percent for the Processes Studied and Sector-Wide................................ 30

Table 6-3. Onsite Practical Minimum Energy Consumption, Energy Savings, and Energy

Savings Percent for the Processes Studied and Sector-Wide................................ 41

Table 7-1. Published Sources Reviewed to Identify Thermodynamic Minimum Energy

Intensities for the Processes Studied ..................................................................... 47

Table 7-2. Thermodynamic Minimum Energy Consumption by Process and Sector-Wide

for the Nine Processes Studied and Extrapolated to Sector Total ........................ 50

Table 8-1. Current Opportunity and R&D Opportunity Energy Savings for the Nine

Processes Studied and Extrapolated to Sector-Wide Total ................................... 52

Table Aa. U.S. Production Volume of Nine Select Petroleum Refining Processes in 2010

with Energy Intensity Estimates and Calculated Onsite Energy

Consumption for the Four Bandwidth Measures (Excludes Feedstock

Energy) .................................................................................................................. 60

Table A2. References for U.S. Throughput Data of the Nine Petroleum Refining

Processes Studied and Energy Intensity Data Used to Calculate the

Current Typical, State of the Art, and Thermodynamic Minimum Energy

Consumption Bands .............................................................................................. 61

Table A3. Technologies Analyzed to Estimate Practical Minimum Energy Intensities............... 63

Table A4. Practical Minimum Technologies Analysis with Weighting Factors .......................... 72

Table A5. Sample Product Fractions from Barrel of Crude Oil Feedstock, Used in

Calculating Distillation Thermodynamic Minimum Energy Intensity ................. 77

Table A6. Change in Gibbs Free Energy for Hydrotreating Reactions ........................................ 79

Table A7. Isomerization Process Stream Assumptions ................................................................ 80

Table A8. Coking Process Assumptions and Calculations ........................................................... 82

Table A9. Calculations Used to Determine Coking Entropy ........................................................ 82

xiv Bandwidth Study on Energy Use and Potential Energy Saving Opportunities in U.S. Petroleum Refining

Bandwidth Study on Energy Use and Potential Energy Saving Opportunities in U.S. Petroleum Refining 1

1. Introduction

1.1. OVERVIEW

This bandwidth study examines energy consumption and potential energy savings opportunities

in the U.S. petroleum refining sector, as defined by classification 324110 of the North American

Industry Classification System (NAICS). The purpose of this data analysis is to provide macro-

scale estimates of energy savings opportunities for petroleum refining processes and petroleum

refining sector-wide. In this study, four different energy consumption bands (or measures) are

estimated. The bandwidth—the difference between bands of energy consumption—is the

estimated potential energy savings opportunity.

The United States produces a wide range of refined petroleum products using multiple processes;

nine of the most energy-intensive petroleum refining processes were studied. Together, these

processes accounted for 68% of the onsite energy consumption of the U.S. petroleum refining

sector in 2010.

The four bands of energy consumption estimated in this report include: the onsite energy

consumption associated with nine petroleum refining processes in 2010 (current typical); two

hypothetical energy consumption levels with progressively more advanced technologies and

practices (state of the art and practical minimum); and one energy consumption level based on

the minimum amount of energy needed to theoretically complete a petroleum refining process

(thermodynamic minimum). The bands of energy consumption are used to calculate current and

R&D opportunity bandwidths for energy savings.

1.2. COMPARISON TO OTHER BANDWIDTH STUDIES

This study builds upon the 2006 DOE bandwidth report Energy Bandwidth for Petroleum

Refining Processes. The earlier study relied on slightly different energy consumption bands.

This study compares diverse industrial, academic and governmental consumption data to

analyses1 of reported plant energy consumption data in the Manufacturing Energy Consumption

Survey (MECS) conducted by the U.S. Energy Information Administration (EIA) for data year

2010. This study also expands the number of petroleum refining processes studied from six to

nine and updates energy consumption and production values to the year 2010.

This report is one in a series of bandwidth studies commissioned by DOE’s Advanced

Manufacturing Office characterizing energy consumption in U.S. manufacturing using a uniform

methodology and definitions of energy bands. Other manufacturing sector bandwidth studies

include chemicals, iron and steel, and pulp and paper; additional sector studies are under

1 The relevant analysis was published as the Manufacturing Energy and Carbon Footprint for the Petroleum

Refining Sector (NAICS 324110), based on energy use data from 2010 EIA MECS (with adjustments) in February

2014. Hereafter, this document will be referred to as the “Energy Footprint” and listed in the References section as

DOE 2014.

2 Bandwidth Study on Energy Use and Potential Energy Saving Opportunities in U.S. Petroleum Refining

consideration. Collectively, these studies explore the potential energy savings opportunities in

manufacturing that are available through existing technology and with investment in research

and development (R&D) technologies.

1.3. DEFINITIONS OF ENERGY CONSUMPTION BANDS AND

OPPORTUNITY BANDWIDTHS

There are four energy consumption bands referenced throughout this report: current typical (CT),

state of the art (SOA), practical minimum (PM), and thermodynamic minimum (TM) energy

consumption. These bands describe different levels of energy consumption for petroleum

refining processes.

As shown in Figure 1-1, the bands

progress from higher to lower levels

of energy consumption, reflecting the

use of increasingly more efficient

manufacturing technologies and

practices. The upper bound is set by a

mix of new and older technologies

and practices in current use (the

current typical level of energy

consumption). The lower bound is

defined by the theoretical minimum

energy requirement assuming ideal

conditions and zero energy losses (the

thermodynamic minimum level of

energy consumption).

Each of these two bounds defining

the extremes of energy consumption

can be compared to hypothetical

measures in the middle of this range.

If manufacturers use the most

efficient technologies and practices available in the world, energy consumption could decrease

from the current typical to the level defined by the state of the art. Since these state of the art

technologies already exist, the difference between the current typical and the state of the art

energy consumption levels defines the current opportunity to decrease energy consumption.

Given that this is an evaluation of technical potential, fully realizing the current opportunity

would require investments in capital that may or not be economically viable for any given

facility.

Widespread deployment of future advanced technologies and practices under investigation by

researchers around the globe could help manufacturers attain the practical minimum level of

Figure 1-1. Energy Consumption Bands and Opportunity

Bandwidths Estimated in this Study


energy consumption. The difference between state of the art and practical minimum levels of

energy consumption defines the R&D opportunity for energy savings.

Definitions of the four energy bands are

provided in the inset (box at right).

Definitions of the two opportunity

bandwidths are provided below:

The current opportunity is the energy

savings that is potentially attainable

through capital investments in the best

technologies and practices available

worldwide. It is the difference between

CT and SOA energy consumption.

The R&D opportunity is the energy

savings that is potentially attainable

through the applied R&D technologies

under development. It is the difference

between SOA and PM energy

consumption. To attain this energy

savings, petroleum refineries would

need to produce refined products in

new ways with technologies that are

not commercially available.

The difference between PM and TM energy consumption is labeled as impractical. The term

impractical is used because with today’s knowledge of technologies in R&D, further investment

may no longer lead to incremental energy savings and thermodynamic limitations impede

technology opportunities.

1.4. BANDWIDTH ANALYSIS METHOD

This Section describes the method used in this bandwidth study to estimate the four bands of

energy consumption and the two corresponding energy savings opportunity bandwidths. This

section can also be used as a guide to understanding the structure and content of this report.

In this study, U.S. energy consumption is labeled as either “onsite energy” or “primary energy”

and defined as follows:

Onsite energy (sometimes referred to as site or end use energy) is the energy consumed

within the manufacturing plant boundary (i.e., within the plant gates). Non-fuel feedstock

energy is not included in the onsite energy consumption values presented in this study.

Definitions of Energy Bands Used in the Bandwidth Studies

The following definitions are used to describe different levels of U.S. energy consumption for a specific manufacturing process industry-wide:

Current Typical (CT) energy consumption: U.S. energy consumption in 2010.

State of the Art (SOA) energy consumption: The minimum amount of energy required assuming the adoption of the best technologies and practices available worldwide.

Practical Minimum (PM) energy consumption: The minimum amount of energy required assuming the deployment of the best applied R&D technologies under development worldwide.

This measure is expressed as a range to reflect the speculative nature of the energy impacts of the unproven technologies considered.

Thermodynamic Minimum (TM) energy consumption: The minimum amount of energy theoretically required assuming ideal conditions typically unachievable in real-world applications.


Primary energy (sometimes referred to as source energy) includes energy that is

consumed both offsite and onsite during the manufacturing process. Offsite energy

consumption includes generation and transmission losses associated with bringing

electricity and steam to the plant boundary. Non-fuel feedstock energy is not included in

the primary energy values. Primary energy is frequently referenced by governmental

organizations when comparing energy consumption across sectors.

Four bands of energy consumption are quantified for select individual processes and petroleum

refining sector-wide. The bands of energy consumption and the opportunity bandwidths

presented herein consider onsite energy consumption; feedstocks2 are excluded. To

determine the total annual onsite CT, SOA, PM, and TM energy consumption values of the

processes studied (TBtu per year), energy intensity values per unit of output or feed (Btu per

barrel) are estimated and multiplied by the output or feed volumes (barrels per year). The year

2010 is used as a base year since it is the most recent year for which consistent sector-wide

energy consumption data are available. Unless otherwise noted, 2010 capacity and throughput or

production data is used. Some petroleum refining processes are exothermic and are net producers

of energy; the net energy was considered in the analysis.

The estimates presented are for macro-scale consideration of energy use in petroleum refining.

The estimates reported herein are representative of average U.S. petroleum refining; they do not

represent energy use in any specific facility or any particular region in the United States or the

world.

Significant investment in technology development and implementation would be needed to fully

realize the potential energy savings opportunities estimated. The costs associated with achieving

SOA and PM energy consumption are not considered in this report; a techno-economic analysis

of the costs and benefits of future R&D technologies was not in the scope of this study.

The calculated energy consumption values in this report are based on an examination of

referenced data and extrapolation to sector-wide energy savings opportunities. The references,

methodology, and assumptions employed are presented with the data in each chapter and were

peer reviewed.

Overview of energy use in petroleum refining: Chapter 2 provides an overview of the U.S.

petroleum refining sector and how energy is used in petroleum refining (how much, what type,

and for what end uses).

Estimating throughput or production volumes for select processes: Chapter 3 presents the

relevant throughput or production volumes for the nine processes (bbl per year) in 2010 and

the rationale for how the nine processes were selected.

2 Feedstock energy is the nonfuel use of combustible energy. Feedstocks are converted to refined petroleum products

(not used as a fuel); MECS values reported as “feedstocks” exclude feedstocks converted to other energy products

(e.g., crude oil converted to residual and distillate fuel oils).


Estimating CT energy consumption: Chapter 4 presents the calculated onsite CT energy

consumption (TBtu per year) for the nine processes individually and sector-wide (along with

references for the CT energy intensity data and assumptions). The CT energy consumption data

is calculated based on this energy intensity data and the throughput or production volumes

(identified in Chapter 3). The boundary assumptions for the industrial processes considered in

this bandwidth study are presented.

MECS provides onsite CT energy consumption data sector-wide for 2010 (See Table 4-1).

However, MECS does not provide CT energy consumption data for individual processes. The

percent coverage of the processes studied (compared to MECS sector-wide data) is presented and

used in calculations discussed later in this report.

Primary CT energy consumption (TBtu per year) estimates are calculated, which include offsite

generation and transmission losses associated with bringing electricity and steam to

manufacturing facilities. Primary energy consumption estimates are not provided for SOA, PM,

or TM because they were outside the scope of this study.

Estimating SOA energy consumption: Chapter 5 presents the estimated onsite SOA energy

consumption for the nine processes (along with the references for the SOA energy intensity data

and assumptions). The sector-wide SOA energy consumption is estimated based on an

extrapolation of the SOA energy consumption for the nine processes studied. The current

opportunity bandwidth, the difference between CT energy consumption and SOA energy

consumption (also called the SOA energy savings), is presented along with the SOA energy

savings percent.

Estimating PM energy consumption: Chapter 6 presents the estimated onsite PM energy

consumption for the nine processes (along with the references for PM energy intensity data and

assumptions). The range of potentially applicable applied R&D technologies to consider in the

PM analysis worldwide is vast. The technologies that were considered are sorted by process and

described in Appendix A3. The technologies that are considered crosscutting throughout all of

petroleum refining along with the most energy-saving, process-specific R&D technologies were

used to determine PM energy consumption for each process. A weighting method that includes

factors such as technology readiness, cost, and environmental impact was developed for all

technologies considered; the weighting analysis methodology and summary table provided in

Appendix A4 is intended to serve as a resource for continued consideration of all identified R&D

opportunities.

The sector-wide PM energy consumption is estimated based on an extrapolation of the PM

energy consumption for the nine processes studied. The R&D opportunity bandwidth, the

difference between SOA energy consumption and PM energy consumption, is presented along

with the PM energy savings percent. PM energy savings is the sum of current and R&D

opportunity.


The technologies considered in the PM analysis are unproven on a commercial scale. As a result,

the PM energy consumption is expressed as a range. The upper limit is assumed to be the SOA

energy consumption; the lower limit is estimated and shown as a dashed line with color fading in

the summary figures because the PM is speculative and depends on unproven R&D technologies.

Furthermore, the potential energy savings opportunity could be greater if additional unproven

technologies were considered.

Estimating TM energy consumption: Chapter 7 presents the estimated onsite TM energy

consumption for the nine processes (along with the references for the TM energy intensity data

and assumptions). The TM energy intensities are based on the commercial process pathways.

TM energy consumption assumes all of the energy is used productively and there are no energy

losses. TM is the minimum amount of energy required; in some cases it is less than zero.

To determine the available potential energy savings opportunities in this bandwidth study, TM

energy consumption was used as the baseline for calculating the energy savings potentials for

each process studied (not zero, as is typically the case in considering energy savings

opportunities). The rationale for using TM as the baseline is explained in Chapter 7.

Estimating the energy savings opportunities: Chapter 8 presents the energy savings

opportunity bandwidths for the processes and sector-wide. The analyses used to derive these

values are explained in Chapters 3 to 7.


2. U.S. Petroleum Refining Sector Overview

This Chapter presents an overview of the U.S. petroleum refining sector, including its impact on

the economy and jobs, number of establishments, types of energy consumed, and the end uses of

the energy. The convention for reporting energy consumption as either onsite versus primary

energy is explained. The data and information in this Chapter provide the basis for understanding

the energy consumption estimates.

2.1. U.S. PETROLEUM REFINING ECONOMIC OVERVIEW

Petroleum refining products are an essential part of the world economy, providing a large source

of energy as well as other high value products. The petroleum refining sector produces many

high value outputs (e.g., gasoline, jet fuel, etc.) that are essential to our economy. Petroleum is

the largest source of primary energy use for the U.S., accounting for 35.3 quadrillion Btu in 2011

(EIA 2012e). About 71% of petroleum is consumed in transportation, making that sector

particularly vulnerable to changes in production and pricing (EIA 2012d). Petroleum refining-

derived products are also relied upon throughout the manufacturing supply chain, both as fuels

and as chemical building blocks in the production of consumer goods.

Petroleum refining plays an outsized role in the U.S. manufacturing sector in terms of the

economy, jobs, and energy. The U.S. was the largest producer of refined petroleum products in

2010, with a net production of a total of over 12.5 million barrels per day (EIA 2013b). The

petroleum refining industry is closely connected to the chemicals industry as certain petroleum

products (e.g., naphtha) are used as feedstocks (or inputs) for the production of petrochemicals,

an important subsector of the chemicals industry. In 2011, the total value of petroleum refining

products shipped in the United States amounted to $753 billion and direct employment of the

petroleum refining sector numbered about 62,000 (USCB 2012b).

2.2. U.S. PETROLEUM REFINING PRODUCTS, ESTABLISHMENTS, AND

PROCESSES

Petroleum refining is a complex industry that generates a diverse slate of fuel and chemical

products, from gasoline to heating oil. Numerous outputs are produced by petroleum refineries

each year in the United States; Table 2-1 shows the type and amount of products produced by

U.S. refineries in 2010.


Table 2-1. U.S. Petroleum Refinery Products, 2010

Product Production

(million barrels)

Distillate Fuel Oil * 1,538

Motor Gasoline 1,142

Jet Fuel 521

Petroleum Coke * 296

Still Gas * 245

Liquefied Refinery Gases (LRG) * 240

Residual Fuel Oil * 210

Asphalt and Road Oil 139

Petrochemical Feedstocks 119

Lubricants 60

Miscellaneous 28

Special Naphthas 14

Kerosene 6

Aviation Gasoline 5

Waxes 3

Total 4,568

Source: EIA 2013b

* A portion of some refining products are used as fuel at the refinery. Approximately 90% of still gas, 28% of petroleum coke, and <1% of distillate fuel oil, residual fuel oil, and LRG are consumed onsite as fuel (EIA 2013f)

As of January 1, 2012, there were 144 operable petroleum refineries (both operating and idle) in

the U.S. (EIA 2012c). About 44% of refineries were located in three states: Texas (which had 27

refineries), Louisiana (19 refineries), and California (18 refineries) (EIA 2012c). Most of the

larger refineries are concentrated along the coast due to the access to sea transportation and

shipping routes. Figure 2-1 shows the current geographic distribution of petroleum refineries in

the U.S. As can be seen, there is a concentration of refineries along the Gulf Coast, in close

proximity to shipping channels and crude oil supply sources.


The refining process involves separating, cracking, restructuring, treating, and blending

hydrocarbon molecules to generate petroleum products. Figure 2-2 shows the typical overall

refining process. The specific processes implemented will vary from refinery to refinery, and

depend upon the refinery’s capacity and feedstock makeup or quality as well as specific

products. As seen in Figure 2-2 petroleum refinery operations generally start with the

atmospheric crude distillation unit. Heavier products from the bottom of the atmospheric

distillation tower continue to the vacuum crude distillation unit where heat is applied under

vacuum conditions to further separate products. To simplify the bandwidth analysis, coking and

visbreaking were combined as one process area, where coking includes both delayed coking and

fluid coking. Also, the isomerization process category includes isobutane, isopentane, and

isohexane production, and the alkylation process includes both sulfuric acid and hydrofluoric

acid catalyst.

Figure 2-1. Geographic Distribution of Petroleum Refineries in the United States

(created from EIA 2013d)


Of the numerous processes involved in petroleum refining, nine were selected for this analysis

on the basis of production volume and energy intensity, including three (coking/visbreaking,

hydrocracking, and isomerization) that were not considered in the previous bandwidth (DOE

2006). This makes the present report more representative of the petroleum refining sector as a

whole. The nine processes studied in this bandwidth analysis are listed in Table 2-2 in

alphabetical order.

Table 2-2. Petroleum Refining Processes Selected

for Bandwidth Analysis

Alkylation

Atmospheric Crude Distillation

Catalytic Hydrocracking

Catalytic Reforming

Coking/Visbreaking

Fluid Catalytic Cracking (FCC)

Hydrotreating

Isomerization

Vacuum Crude Distillation

Figure 2-2. Typical Petroleum Refinery Flow Diagram (DOE 2007)


2.3. U.S. PETROLEUM REFINING ENERGY CONSUMPTION

Onsite energy and primary energy for the U.S. petroleum refining sector are provided in Table

2-3. EIA MECS provides onsite energy consumption data by end use, including onsite fuel and

electricity consumption, as well as feedstock energy. Primary energy includes assumptions for

offsite losses (DOE 2014).

Table 2-3. U.S. Petroleum Refining Energy Consumption Sector-Wide, 2010

Onsite Energy Consumption

(includes electricity, steam, and fuel energy used onsite at the facility) 3,176 TBtu

Primary Energy Consumption

(includes onsite energy consumption, and offsite energy losses associated with generating

electricity and steam offsite and delivering to the facility)

3,555 TBtu

Source: DOE 2014

Petroleum refining is the second largest consumer of energy in U.S. manufacturing, accounting

for 3,555 TBtu (18%) of the 19,237 TBtu of total primary manufacturing energy consumption in

2010 (DOE 2014). Offsite electricity and steam generation and transmission losses in petroleum

refining totaled 379 TBtu in 2010; onsite energy consumed within the boundaries of U.S.

petroleum refineries totaled 3,176 TBtu.

Figure 2-3 shows the total onsite energy entering U.S. petroleum refineries; most of the energy

entering is in the form of fuel. About 30% of this fuel is used onsite in boilers and combined heat

and power (CHP) to generate additional electricity and steam (DOE 2014). In contrast, Figure

2-4 shows the total onsite energy at the point of end use. Electricity and steam from both offsite

and onsite generation are included in Figure 2-4, along with the portion of energy loss that

occurs in onsite generation. The data provided in Table 2-3, Figure 2-3, and Figure 2-4 are based

on MECS with adjustments to account for withheld and unreported data (DOE 2014).


Figure 2-3. Onsite Energy Entering U.S. Petroleum Refineries, 2010 (DOE 2014)

Figure 2-4. Onsite Energy Consumption at Point of End Use in U.S. Petroleum Refineries, 2010 (DOE 2014)


2.3.1. Fuel and Feedstocks

As shown in Figure 2-3, onsite fuel consumption amounted to 2,873 TBtu in 2010, or about 90%

of total onsite energy entering petroleum refineries (EIA 2013). Still gas (also referred to as

refinery fuel gas or waste gas) accounts for nearly half (46%) of the sector’s fuel consumption.

Still gas, the most commonly used byproduct fuel, is produced and captured in process off-gases

and is typically made up of hydrogen, methane, ethane, and other light-end gases. The hydrogen

content in still gas improves the enthalpy of combustion which allows for greater transfer of heat

into the process compared to purchased natural gas.

Many refineries are self-sufficient in fuel consumption, using a mixture of byproduct off-gases

and fuel oil. In some cases, however, it is more profitable to process the gas-oil fractions further

into higher-value refined products than to burn them as fuel oil and purchase cheaper natural gas

(and sometimes coal) to supplement the available off gases. Figure 2-5 provides a breakdown of

fuel consumption in the petroleum refining sector by end use in 2010. The categories of end use

are reported by EIA in MECS. The majority of the fuel (64%) is used directly for process

heating; some examples of process heating equipment include fired heaters, heated reactors, and

heat exchangers. A significant portion of fuel (30%) is used indirectly in boilers and CHP to

generate additional onsite electricity and steam (DOE 2014).

The process heating end use is limited to direct-fired fuel used in furnaces for distillation

columns, reactors, and hot oil loops. The indirect boiler and CHP end use includes fuel used for

steam and electricity generation. The values in Figure 2-5 represent total fuel consumption, not

absorbed duty in the process, and therefore boiler, furnace, and turbine efficiency effects are

already accounted for.


Feedstock energy is the nonfuel use of combustible energy. For petroleum refining, feedstock

energy is converted to products instead of being used as a fuel. For the energy bandwidth study,

only the fuel use of combustible energy is considered in the opportunity analysis; however, due

to the highly connected nature of feedstock and fuel energy, it is important to provide some

context around that relationship and some background information on feedstock energy in this

sector. In MECS, EIA reports feedstock energy used for the production of non-energy products

(e.g., asphalt, waxes, lubricants) and as a petrochemical feedstock only; feedstock energy used

for the production of other energy products (e.g., motor gasoline, fuel oil) is excluded. A detailed

breakdown of feedstock by energy type is not available from MECS; instead, all feedstock

energy is classified as ‘other’ fuel. EIA excludes inputs and feedstock that were converted to

other types of energy products from this reported feedstock number, so total feedstock energy

consumed in refining is significantly greater than 2,746 TBtu when accounting for these

additional feedstocks (e.g., crude oil converted into gasoline).

Figure 2-5. Fuel Consumption in the Petroleum Refining Sector by End Use, 2010 (DOE 2014)

Feedstock energy is a significant portion of energy consumption in U.S. petroleum refining

(2,746 TBtu) so it is important to consider when analyzing overall petroleum refining sector

energy use. However, feedstock energy is not included in the onsite energy data in the

energy consumption bands in this study. Feedstock energy is excluded in order to be

consistent with previous bandwidth studies and because the relative amount of feedstock energy

versus fuel energy used in manufacturing is not readily available for individual processes.


2.3.2. Electricity

Figure 2-3 shows that onsite net electricity entering petroleum refineries totaled 153 TBtu in

2010. The data presented is the net amount, which is the sum of purchases and transfers from

offsite sources as well as generation from non-combustion renewable resources (e.g.,

hydroelectric, geothermal, solar, or wind energy) less the amount of electricity that is sold or

transferred out of the plant. Figure 2-4 shows that 213 TBtu of total electricity is consumed at the

point of end use and includes 60 TBtu of electricity generated onsite.

In Figure 2-6, the breakdown of the 213 TBtu of electricity is shown by end use in 2010 (DOE

2014). There are numerous uses for electricity in petroleum refining; the most common use is for

machine driven equipment (i.e., motor-driven systems such as compressors, fans, pumps, and

materials handling and processing equipment). Motors used for cooling water circulation pumps

and fans, however, are accounted for in process cooling end use. Other end uses of electricity for

petroleum refining are less significant, but include nonprocess facility related end uses (e.g.,

facility heating, ventilation, and air conditioning (HVAC), facility lighting, cooking, office

equipment, etc.) and other end uses.

Figure 2-6. Electricity Consumption in the Petroleum Refining Sector by End Use, 2010 (DOE

2014)


2.3.3. Steam

Figure 2-3 shows 150 TBtu of net steam entering petroleum refineries in 2010. The data

presented is the net amount, which is the sum of purchases, generation from renewables, and net

transfers. A larger amount of steam is generated onsite. Figure 2-4 shows that 748 TBtu of steam

is consumed at the point of end use, including 598 TBtu of steam generated onsite (150 TBtu of

purchased and generated steam is lost through distribution to end uses) (DOE 2014).

Figure 2-7 shows the breakdown of 589 TBtu of steam by end use in 2010 (DOE 2014). A

majority of the offsite- and onsite-generated steam is used for process heating; other end uses for

steam in petroleum refining include machine driven equipment (i.e., steam turbines), other

process end uses, and process cooling and refrigeration. Unlike fuel and electricity end use,

steam end use is not reported in MECS. The end use distribution shown here was determined in

the Energy Footprint analysis (DOE 2014) based on input from an industry-led working group.

Figure 2-7. Steam Consumption in the Petroleum Refining Sector by End Use, 2010 (DOE 2014)


3. Throughputs and Production in U.S.

Petroleum Refining

In this bandwidth study, nine petroleum refining processes were selected for individual analysis.

For petroleum refining, process data is often reported in terms of capacity, throughput, or

production. Capacity is the maximum amount of feed that a petroleum refining process unit is

able to process. Throughput is the amount of feed that is actually processed, and is the value used

for energy consumption calculations for seven of the processes in this study. Two of the

processes (alkylation and isomerization) instead use production values of the process product(s).

Table 3-1 presents the U.S. relevant capacity and throughput or production data for the nine

processes for the year 2010.

Table 3-1. U.S. Petroleum Refining Process Capacities and Throughputs/Production for the Nine

Processes Studied in 2010

Process 2010 Capacity

(Million barrels/year)

2010 Throughput or Production*

(Million barrels/year)

Details

Alkylation 423 365** Based on alkylate production capacity

Atmospheric Crude Distillation N/A 5,540 Gross input to atmospheric distillation units (considered equivalent to throughput)

Catalytic Hydrocracking 616 532** Based on downstream charge capacity

Catalytic Reforming 1,221 1,055** Based on downstream charge capacity

Coking/Visbreaking 892 770** Based on downstream charge capacity

Fluid Catalytic Cracking 2,115 1,827** Based on downstream charge capacity

Hydrotreating 5,589 4,829** Based on downstream charge capacity

Isomerization 235 203** Based on isobutane, isopentane, and isohexane production capacity

Vacuum Crude Distillation 2,898 2,504** Based on downstream charge capacity

* Values for alkylation and isomerization are production; all other process values are throughput

**Capacity factor of 86.4% for 2010 (EIA 2011) applied to capacity determine throughput/production

Sources: Appendix A2 provides the sources referenced for U.S. capacity and throughput/production data of each process.

The most energy intensive processes were selected for this study. Other, less intensive processes

were added to the study to expand the representative coverage of the petroleum refining sector as

a whole. In general, the selection of processes was largely dependent on the availability of

current production and energy consumption data.

The year 2010 was used for throughput or production and capacity values to correspond with the

latest MECS data, which is also for 2010. In the data provided by EIA, capacity is either listed


for the year (2010) or at the start of the following year (January 1, 2011). Production data was

gathered from a variety of EIA data sources, including the downstream charge capacities of

operable petroleum refineries, production capacities of operable petroleum refineries, and

refinery utilization and capacity. Throughput for atmospheric crude distillation was directly

available from the data sources, while throughputs or production for the other processes studied

were calculated based on the petroleum refinery utilization rate of 86.4% for 2010. A detailed

listed of sources can be found in Appendix A2.

The total crude distillation rate of all the refineries in the U.S. was about 15 million barrels per

stream day (BPSD) in 2010, versus a capacity of 17.6 million BPSD (EIA 2013c). The crude

distillation capacity of individual refineries varies widely from about 3,300 BPSD to 625,000

BPSD (EIA 2012d). The U.S. Small Business Administration classifies refineries with a crude

distillation capacity of less than 125,000 barrels per day as small and those with more than

125,000 barrels per day as large (SBA 2012). Although there are more small refineries than large

ones, they only account for about 25% of total U.S. refining capacity.

Refinery size has an impact on operating practices and energy efficiency. Typically, small

refineries are less complex than medium and large refineries and typically have fewer of the

refining process units listed in Figure 2-2. On the other hand, refineries larger than 200,000

barrels per day often have more than one parallel train for the same process units (i.e., two crude

distillation towers or two reformers) as a consequence of staged refinery expansions over time

(most refineries have a service life of 50-100 years).


4. Current Typical Energy Consumption for U.S.

Petroleum Refining

This Chapter presents the energy consumption data for individual petroleum refining processes

and sector-wide in 2010. Energy consumption in a manufacturing process can vary for diverse

reasons. The energy intensity estimates reported herein are representative of average U.S.

petroleum refining; they do not represent energy consumption in any specific facility or any

particular region in the United States.

4.1. BOUNDARIES OF THE PETROLEUM REFINING BANDWIDTH STUDY

Estimating energy requirements for an industrial process depends on the boundary assumptions;

this is especially true in the petroleum refining industry. The key focus of this bandwidth study is

energy consumption within the plant boundary, which is the onsite use of process energy

(including purchased energy and onsite generated steam and electricity) that is directly applied to

petroleum refining.

This study does not consider lifecycle energy consumed during raw material extraction (e.g.,

drilling and exploration), off-site treatment, and transportation of materials. Upstream energy,

such as the energy required for processing and handling materials outside of the plant is also not

included. To be consistent with previous bandwidth studies, feedstock energy and the energy

associated with delivering feedstocks to the plant gate (e.g., producing, conditioning, and

transporting feedstocks) are excluded from the energy consumption bands in this analysis.

4.2. ESTIMATED ENERGY INTENSITY FOR INDIVIDUAL PROCESSES

Energy intensity data are needed to calculate bands of energy consumption in this study. This

Section presents the estimated energy intensities of the nine processes studied.

The specific energy needed to process a barrel of feed or produce a barrel of product can vary

significantly between processes, and also between facilities. Energy intensity is a common

measure of energy performance in manufacturing. Energy intensity is reported in units of energy

consumption (typically Btu) per unit of feed or output (typically barrels in the petroleum refining

industry) and, therefore, reported as Btu per barrel (Btu/bbl). Energy intensity estimates are

available for specific equipment performance, process unit performance, or even plant-wide

performance. Energy intensity can be estimated by process, both in the United States and other

global regions, based on average, representative process and plant performance.

Energy efficiency at petroleum refineries varies depending on regional supply and demand

constraints, feedstock characteristics, facility and equipment age, weather conditions, and

applicable compliance requirements. Energy efficiency also varies within an individual refinery

for many of the same reasons. Swings of 10-15% in energy efficiency are not uncommon

whether due to external or internal factors (Kumana & Associates 2013). External factors include


variations in feedstock flow, feedstock properties (due to blending), feedstock supply

temperature, ambient temperature, and market demand for refined products. Internal factors all

stem from either poor design or poor operating practices, the worst of which are usually

inefficient heat exchanger network (HEN) design, poor process control due to inadequate

instrumentation, failure to reconcile measured data, operating equipment at off-design

conditions, unscheduled equipment outages, and sub-optimal heat exchanger cleaning programs

(Kumana & Associates 2013).

Appendix A1 presents the CT energy intensities and energy consumption for the nine processes

studied in alphabetical order. Table 4-1 presents a summary of the references consulted to

identify CT energy intensity by process. Appendix A2 provides the references used for each

process.

Because the petroleum refining sector is diverse, covering many products, a range of data

sources were considered (see Table 4-1). Current energy intensities for each of the nine

petroleum refining processes identified in this study were taken from the Energy and

Environmental Profile of the U.S. Petroleum Refining Industry, denoted in this report as DOE

2007. Energy intensity values for each of the processes are broken down by electricity and

steam/process heat consumption in DOE 2007, resulting in net energy intensity. This study

provided the most current representative values for the U.S. petroleum refining industry. Sources

for energy intensities in DOE 2007 came from a variety of sources, including Hydrocarbon

Processing’s 2006 Refining Processes Handbook. This commonly referenced industry

publication provides utility requirements for licensed technologies currently used in petroleum

refineries.

Since the publishing of DOE 2007, two additional Refining Processes Handbooks (2008 and

2011) have been released. These sources were reviewed and it was determined that they did not

provide new information for the nine selected processes. As a result, the energy intensity values

are taken directly from DOE 2007. Other sources such as Solomon Associates’ Comparative

Performance Analysis (well known in the petroleum refining sector for their performance

monitoring and benchmarking services) were sought but ultimately unavailable due to

confidentiality considerations.

The specific technologies and processes in use at individual petroleum refineries can vary; thus,

it is difficult to ascertain an exact amount of energy necessary to complete a process.

Consequently, the values for energy intensity provided should be regarded as estimates based on

the best available information.


Table 4-1. Published Sources Reviewed to Identify Current Typical Energy Intensities for Processes

Studied

Source Description

DOE 2007

The Energy and Environmental Profile of the U.S. Petroleum Refining Industry,

published by DOE in 2007 and focused on the U.S., provides a detailed energy breakdown (including total net process energy) for each of the petroleum refining processes that consume a significant amount of energy in the sector and produce numerous products. This was the main source for determining current typical energy intensity.

EIA 2013a

Manufacturing Energy Consumption Survey data released by EIA every four years; this data comes from a survey that is taken by U.S. manufacturers. The most recent year for which MECS data is published is 2010. The data is scaled up to cover the entirety of U.S. manufacturing and for individual manufacturing sectors. For petroleum refining, it provides energy consumption data for the entire sector.

HP 2011, HP 2008a, HP 2006

Hydrocarbon Processing publishes the Refining Processes Handbook every few years (including the 2006, 2008, and 2011 versions that were consulted for this bandwidth) which provides utility consumption data (electricity, steam, and fuel) for specific licensed petroleum refining processing technologies. In some cases, multiple technologies are listed for a specific process, allowing for comparison of energy consumption data.

4.3. CALCULATED CURRENT TYPICAL ENERGY CONSUMPTION FOR

INDIVIDUAL PROCESSES

Table 4-2 presents the calculated onsite CT energy consumption for the nine processes studied.

To calculate onsite CT energy consumption, energy intensity for each process (presented initially

in Appendix A1) is multiplied by the 2010 throughput or production data (presented initially in

Table 3-1 and also in Appendix A1). Feedstock energy is excluded from the consumption values.

The CT energy consumption for these nine processes is estimated to account for 2,163 TBtu of

onsite energy, or 68% of the 3,176 TBtu of sector-wide onsite energy use in 2010. Appendix A1

presents the onsite CT energy consumption for the nine processes individually in alphabetical

order.

Calculated primary CT energy consumption by process is also reported in Table 4-2. Primary

energy includes offsite energy generation and transmission losses associated with electricity and

steam from offsite sources. To determine primary energy, the net electricity and net steam

portions of sector-wide onsite energy are scaled to account for offsite losses and added to onsite

energy (see the footnote in Table 4-2 for details on the scaling method).


Table 4-2. Calculated U.S. Onsite Current Typical Energy Consumption for Processes Studied in 2010 with

Calculated Primary Energy Consumption and Offsite Losses

Process

Energy Intensity

(Btu/bbl)

Throughput or Production

1

(million bbl/year)

Onsite CT Energy

Consumption, Calculated

(TBtu/year)

Offsite Losses,

Calculated

(TBtu/year)2

Primary CT Energy

Consumption, Calculated

(TBtu/year)

Alkylation 246,700 365 90 11 101

Atmospheric Crude

Distillation 109,100 5,540 604 72 676

Catalytic Hydrocracking 158,900 532 85 10 95

Catalytic Reforming 263,900 1,055 279 33 312

Coking/Visbreaking 147,700 770 114 14 127

Fluid Catalytic Cracking 182,800 1,827 334 40 374

Hydrotreating 80,800 4,829 390 46 437

Isomerization 216,000 203 44 5 49

Vacuum Crude Distillation 89,100 2,504 222 27 250

Total for Processes Studied

2,163 257 2,420

Current typical (CT) 1 Values for alkylation and isomerization are production; all other process values are throughput

2 Accounts for offsite electricity and steam generation and transmission losses. Offsite electrical losses are based on published

grid efficiency. EIA Monthly Energy Review, Table 2.4, lists electrical system losses relative to electrical retail sales. The energy

value of electricity from offsite sources including generation and transmission losses is determined to be 10,553 Btu/kWh. Offsite

steam generation losses are estimated to be 20% (Swagelok Energy Advisors, Inc. 2011. Steam Systems Best Practices) and

offsite steam transmission losses are estimated to be 10% (DOE 2007, Technical Guidelines Voluntary Reporting of Greenhouse

Gases and EPA 2011, ENERGY STAR Performance Ratings Methodology).

References for throughput or production data and energy intensity data are provided by process in Appendix A2. The other

values are calculated as explained in the text.

http://www.swagelok.com/Chicago/Services/Energy-Services/~/media/Distributor%20Media/C-G/Chicago/Services/ES%20-%20Thermal%20Cycle%20Efficiency_BP_33.ashx

http://www.eia.gov/oiaf/1605/January2007_1605bTechnicalGuidelines.pdf


http://www.energystar.gov/ia/business/evaluate_performance/site_source.pdf


4.4. CURRENT TYPICAL ENERGY CONSUMPTION BY PROCESS AND

SECTOR-WIDE

In this Section, the CT energy consumption estimates for nine processes studied are provided.

Table 4-3 presents the onsite CT energy consumption by process and sector-wide for U.S.

petroleum refining. The nine processes studied account for 68% of all onsite energy consumption

by the U.S. petroleum refining sector in 2010. As shown in the last column of Table 4-3, the

percentage of coverage of the processes studied is calculated. This indicates how well the

processes studied represent total sector-wide MECS-reported energy. The overall percentage of

coverage for the processes studied (68%) is used later in this study to determine the extrapolated

total sector-wide SOA, PM, and TM energy consumptions.

Table 4-3 also presents CT primary energy consumption by process. Primary energy is calculated

from onsite CT energy consumption databased on an analysis of MECS data (DOE 2014), with

scaling to include offsite electricity and steam generation and transmission losses (DOE 2014).


Table 4-3. Onsite and Primary Current Typical Energy Consumption for the Nine Processes Studied

and Sector-Wide in 2010, with Percent of Sector Coverage

Process

Onsite CT Energy

Consumption,

calculated

(TBtu/year)

Primary CT Energy

Consumption,

calculated*

(TBtu/year)

Percent Coverage

(Onsite CT as a % of

Sector-wide Total)**

Alkylation 90 101 19%

Atmospheric Crude Distillation 604 676 7%

Catalytic Hydrocracking 85 95 9%

Catalytic Reforming 279 312 12%

Coking/ Visbreaking 114 127 3%

Fluid Catalytic Cracking 334 374 11%

Hydrotreating 390 437 4%

Isomerization 44 49 3%

Vacuum Crude Distillation 223 250 1%

Total for Processes Studied 2,163 2,420 68%

All Other Processes 1,104 1,135 32%

Total for Petroleum Refining Sector-wide

3,176*** 3,555*** 100%

Current typical (CT)

* Accounts for offsite electricity and steam generation and transmission losses. Offsite electrical losses are based on published grid efficiency. EIA Monthly Energy Review, Table 2.4, lists electrical system losses relative to electrical retail sales. The energy value of electricity from offsite sources including generation and transmission losses is determined to be 10,553 Btu/kWh. Offsite steam generation losses are estimated to be 20% (Swagelok Energy Advisors, Inc. 2011. Steam Systems Best Practices) and offsite steam transmission losses are estimated to be 10% (DOE 2007, Technical Guidelines Voluntary Reporting of Greenhouse Gases and EPA 2011, ENERGY STAR Performance Ratings Methodology).

** Calculated by dividing the onsite CT energy consumption for the processes studied by sector-wide onsite CT energy consumption (3,176 TBtu).

*** Source for sector-wide values is DOE 2014.





http://www.energystar.gov/ia/business/evaluate_performance/site_source.pdf


5. State of the Art Energy Consumption for U.S.

Petroleum Refining

As plants age, manufacturing processes and equipment are updated and replaced by newer, more

energy-efficient technologies. This results in a range of energy intensities among U.S. petroleum

refineries. Petroleum refineries will output a wide range of refined products using a different

series of processes and types of feedstocks, and will therefore vary widely in size, age,

efficiency, and energy consumption. Modern petroleum refineries can benefit from more energy-

efficient technologies and practices.

A modern refinery is designed for flexibility in the type and amount of feedstock that is available

as well as the desired outputs. The energy consumption necessary to produce a higher quality

product often depends on the amount and type of processing the feedstock crude oil must

undergo. Often, refineries will elect to process the most readily available or economical

feedstock to maximize productivity. Energy consumption often may be a secondary

consideration as refineries choose modifications to increase output and thus profit. Instead of

performing costly equipment replacement that may interrupt operations, refineries may be more

inclined to seek ways to optimize current operations, investing in improvements to existing

equipment, taking advantage of heat or power recovery, or adopt an energy management system.

This Chapter estimates the energy savings possible if U.S. petroleum refineries adopt the best

technologies and practices available worldwide. State of the art (SOA) energy consumption is the

minimum amount of energy that could be used in a specific process using existing technologies

and practices.

5.1. CALCULATED STATE OF THE ART ENERGY CONSUMPTION FOR


Appendix A1 presents the onsite SOA energy intensity and consumption for the nine processes

considered in this bandwidth study in alphabetical order. The SOA energy consumption for each

petroleum refining process is calculated by multiplying the SOA energy intensity for each

process by the relevant throughput or production (all relevant data are presented in Appendix

A1).

The onsite SOA energy consumption values are the net energy consumed in the process using the

single most efficient process and production pathway. No weighting is given to processes that

minimize waste, feedstock streams, and byproducts, or maximize yield, even though these types

of process improvements can help minimize the energy used to produce a pound of product. The

onsite SOA energy consumption estimates exclude feedstock energy.

Table 5-1 presents the published sources referenced to identify the SOA energy intensities.

Technologies identified that are in a pre-commercial stage of development or that are extremely


expensive were not considered in the SOA analysis (instead they were considered in Chapter 6

on the practical minimum (PM) energy consumption).

Table 5-1. Published Sources Referenced to Identify State of the Art Energy Intensities for Nine Select

Processes

Source Abbreviation Description

DOE 2013

Compiled results from the 14 DOE Energy Saving Assessments that were conducted

at petroleum refineries during 2006-2011 by DOE and Oak Ridge National Laboratory

to determine average onsite energy savings.

ExxonMobil 2011

Compiled results from the 14 DOE Energy Saving Assessments that were conducted

at petroleum refineries during 2006-2011 by DOE and Oak Ridge National Laboratory

to determine average onsite energy savings.

Kumana & Associates

2013

The data represent results from 22 case studies for seven types of process units, both

existing and new. The studies were conducted on nine refineries in the U.S., China,

and the Middle East between 2003 and 2013.

LBNL 2005a This 2005 study reviews energy efficiency opportunities available for petroleum

refineries.

McKinsey 2009 This report discusses methods to reduce greenhouse gas emissions and energy

consumption for 10 different sectors, including petroleum refining.

SEP 2013

The Superior Energy Performance (SEP) Program is currently being demonstrated in

numerous manufacturing facilities, 13 of which have achieved certification with various

levels of realized percent energy savings. Achieving SEP involves a rigorous

certification including adopting an energy management system and measurement and

verification of continuous energy improvement.

Other various articles and reports; see Appendix A2 for full references for each process

An important and reliable source of actual field data for SOA energy intensity was from 22

energy efficiency studies completed between 2003 and 2013 at nine different refineries located

in China, the Middle East, and the U.S.3 Energy efficiency improvements were identified for

multiple refining processes (integrated atmospheric and vacuum distillation, reforming,

hydrotreating, delayed coking, hydrocracking, and isomerization). Significant savings potential

was found, ranging from 10-33% at typical simple paybacks of less than five years (see Table

5-2). The primary technique employed was heat recovery optimization using pinch analysis, but

included some equipment upgrades (e.g., variable frequency drives for selected large motors),

minor process modifications, operational best practices such as load management, and

optimizing the design and operation of the site CHP systems. Although most of the efficiency

improvements involved retrofitting existing process units, significant savings were also

identified in licensor design offerings for completely new units. All the savings projects were

identified through application of pinch analysis for optimized heat recovery.

3 Jimmy Kumana, a well-respected petroleum refining industry consultant and energy efficiency course instructor,

agreed to serve as a peer reviewer of this petroleum refining bandwidth study and to share the results of case study

findings from petroleum refining energy audits conducted in the U.S., China, and the Middle East. The energy

savings data provided by his company was consolidated and findings were reported by year and process units only.

Information linking their findings with specific refineries has been deliberately omitted.


Table 5-2. Consulted Petroleum Refining SOA Case Studies Breakdown and Details

Process Unit Number of

Units

% Savings

Identified

Integrated Atmospheric and Vacuum Crude Distillation

6 17%

Coking/Visbreaking 2 15%

Hydrocracking 2 20%

Hydrotreating 6 33%

Isomerization 2 10%

Reforming 4 15%

Source: Kumana & Associates 2013

These savings can be considered to correspond to SOA designs, using existing catalysts and

distillation conditions, and taking into account constraints of safety, operability, and layout. The

savings potential generally ranged from 10% to 20%, but was over 30% for hydrotreating.

Achieving greater savings would likely only be possible by switching to newer and better (more

expensive) catalysts that can deliver higher yields, faster kinetics, and greater resistance to

activity loss over time. In the crude distillation unit (CDU), improved heat recovery is often

limited because of accelerated asphaltene fouling rates at higher temperatures. The solution is to

employ a multi-pronged strategy that includes optimized feedstock blending, raising desalter and

flash drum temperatures, mechanical modifications to the heat exchanger to ensure adequate

fluid velocities and shear rates, and optimized heat exchanger cleaning strategies.

Although all the identified projects in the case studies were both technically and economically

feasible, not all were implemented for a variety of reasons including a strategic decision to

withhold investment in an urban refinery where a shutdown was being considered, competition

for limited capital from higher priority projects (e.g., capacity, safety, environmental), disputes

with the licensor over intellectual property rights for the improved design, and inability to

incorporate design revisions into the engineering, procurement, and construction (EPC) schedule.

Another important source of data was a comprehensive report on ways for petroleum refineries

to increase their energy efficiency by Lawrence Berkeley National Laboratory (LBNL) is Energy

Efficiency Improvement and Cost Saving Opportunities for Petroleum Refineries (LBNL 2005a).

This report provides a detailed and valuable overview of energy efficiency opportunities for

specific process units as well as in general for petroleum refineries.

The availability of an estimate of energy savings for each measure varies, and it is important to note

that all the improvement opportunities will likely not be able to be adopted by an individual refinery

due to savings overlap. The authors of this report chose a combination of savings estimates, including

adoption of process control, steam distribution savings from improvements, process integration, and

the use of power recovery. A general listing of energy efficiency opportunity areas for specific

process units as outlined by LBNL 2005a is included in Table 5-3 below.


Table 5-3. Specific Petroleum Refining Process Unit Energy Efficiency Opportunities*

Process Unit Opportunity Areas

Alkylation Process controls, process integration (pinch analysis), optimization distillation

Atmospheric Crude

Distillation

Process controls, high-temperature combined heat and power (CHP), process

integration (pinch analysis), furnace controls, air preheating, progressive crude

distillation, optimization distillation

Coking/Visbreaking Process integration (pinch analysis), furnace controls, air preheating, optimization

distillation

Fluid Catalytic Cracking Process controls, power recovery, process integration (pinch analysis),furnace

controls, air preheating, optimization distillation, process flow changes

Hydrocracking Power recovery, process integration (pinch analysis),furnace controls, air

preheating, optimization distillation

Hydrotreating Process controls, process integration (pinch analysis), optimization distillation,

new hydrotreater designs

Reforming Process integration (pinch analysis), furnace controls, air preheating, optimization

distillation

Vacuum Crude Distillation Process controls, process integration (pinch analysis), furnace controls, air

preheating, optimization distillation

* Adapted from LBNL 2005a

Other valuable sources provided estimates of general savings for SOA, including actual plant

assessments returned from the DOE’s Energy Savings Assessment (ESA) program and

ExxonMobil’s GEMS (Global Energy Management System) studies. The ESA program

identified an aggregate onsite energy savings of about 3.4% for 14 which were surveyed between

2006 and 2011. These ESA analyses identified a number of energy savings opportunities.

Generally, most of the opportunities were related to management of steam systems or process

heating. The targeted steam system improvements related to reducing leaks, adjusting the boilers,

improving insulation and improving steam traps. The other large source of savings came from

improving boilers by upgrading the heat exchangers or by increasing their efficiency. Many of

these simple yet straightforward changes result in savings over CT with unreasonable capital

investment and short payback periods (less than 2 years). However, the ESA savings are

comparatively small compared to other savings estimates (potentially even insignificant when

considering the variability and typical swings in plant energy consumption).

ExxonMobil’s GEMS studies identified savings of about 9% in their petroleum refining

operations due to improvements made between 2002 and 2011 (ExxonMobil 2011). These

savings were applied to each of the nine process units selected for this study4. The McKinsey &

Company report Pathways to a Low-Carbon Economy also estimated that the entire petroleum

refining industry could reduce its energy consumption by 13% over CA by 2020 if it adopted

currently available best practice technologies (McKinsey 2009). The improvements identified by

4 Savings were applied to 2010 energy consumption, even though ExxonMobil findings considered 2002 as the base

year. Nine percent savings were determined to be in line with other SOA sources.


McKinsey 2009 include waste-heat recovery through heat integration, replacing boilers, heaters,

turbines, and motors, and adopting increased cogeneration.

5.2. STATE OF THE ART ENERGY CONSUMPTION BY PROCESS AND

SECTOR-WIDE

Table 5-4 presents the onsite SOA energy consumption for the nine U.S. petroleum refining

processes studied. In this table, the SOA energy consumptions for the processes studied are

summed and extrapolated to provide a sector-wide onsite SOA energy consumption. Table 5-4

also presents the onsite SOA energy savings, or the current opportunity. The SOA energy

savings is also expressed as a percent in Table 5-4. This is also shown in Figure 5-1. It is useful

to consider both TBtu energy savings and energy savings percent when comparing the energy

savings opportunity. Both are good measures of opportunity; however, the conclusions are not

always the same. In Figure 5-1, the percent savings is the percent of the overall energy

consumption bandwidth, with CT energy consumption as the upper benchmark and TM as the

lower baseline. In Figure 5-2, the current energy savings opportunity is shown in terms of

TBtu/year savings for each process. The pie chart in Figure 5-2 captures the blue portions of the

bar chart shown in Figure 5-1. Among the processes studied, the greatest current opportunity in

terms of percent energy savings is vacuum crude distillation at 23% energy savings; the greatest

current opportunity in terms of TBtu savings is atmospheric crude distillation at 82 TBtu per

year savings.

To extrapolate the sector-wide data presented in Table 5-4 and Figure 5-1, the SOA energy

consumption of each individual process studied is summed, and the sum is divided by the percent

coverage for the entire subsector (68%, as shown in Table 4-3). Percent coverage is the ratio of

the sum of all the CT energy consumption for the individual processes studied to the CT energy

consumption for the subsector provided by MECS (see Table 4-3). The extrapolated number is

the estimated SOA energy consumption for the entire subsector. The SOA energy consumption

for the remainder of the sector (i.e., all processes that are not included in the nine processes

studied) was calculated by subtracting the total for the processes studied from the sector-wide

total. These additional processes are together referred to as All Other Processes in Table 5-4).

Table 5-4 also presents the SOA energy savings percent. To calculate the onsite SOA energy

savings percent, the thermodynamic minimum (TM) energy consumption serves as the baseline

for estimating percent energy savings, not zero. The energy savings percent is the percent of

energy saved with SOA technologies and practices compared to CT energy consumption,

considering that the TM may not be zero. As will be explained in Chapter 7, the TM reaction

energy for some petroleum refining processes is a negative value. When comparing energy

savings percent from one process to another, the absolute savings is the best measure of

comparison. The equation for calculating onsite SOA energy savings percent is:

𝑆𝑂𝐴 𝑆𝑎𝑣𝑖𝑛𝑔𝑠 % = 𝐶𝑇 − 𝑆𝑂𝐴

𝐶𝑇 − 𝑇𝑀


Table 5-4. Onsite State of the Art Energy Consumption, Energy Savings, and Energy Savings Percent for the

Processes Studied and Sector-Wide

Process

Onsite CT

Energy

Consumption

(TBtu/year)

Onsite SOA

Energy

Consumption

(TBtu/year)

SOA Energy

Savings†

(CT-SOA)

(TBtu/year)

SOA Energy

Savings

Percent

(CT-SOA)/

(CT-TM)*

Alkylation 90 80 10 9%

Atmospheric Crude Distillation 604 522 82 20%

Catalytic Hydrocracking 85 74 11 9%

Catalytic Reforming 279 246 33 17%

Coking/ Visbreaking 114 101 12 11%

Fluid Catalytic Cracking 334 289 45 17%

Hydrotreating 390 333 57 9%

Isomerization 44 39 5 11%

Vacuum Crude Distillation 223 192 32 23%

Total for Processes Studied 2,163 1,877 286 14%

All Other Processes 1,013 879 134 14%

Total for Petroleum Refining Sector-

wide 3,176 2,756** 420 14%

Current typical (CT), State of the art (SOA)

† SOA energy savings is also called Current Opportunity.

* SOA energy savings percent is the SOA energy savings opportunity from transforming petroleum refining processes. Energy savings

percent is calculated using TM energy consumption shown in Table 7-2 as the minimum energy consumption. The energy savings

percent, with TM as the minimum, is calculated as follows: ( CT- SOA)/(CT- TM)

** The sector-wide SOA energy consumption was an extrapolated value, calculated by dividing the total onsite SOA energy

consumption for the processes studied by the overall percent coverage from Table 4-3 (68%).


Figure 5-1. Current Opportunity Energy Savings Bandwidths for Processes Studied (with Percent of

Overall Energy Consumption Bandwidth)


5.2.1. Comparing State of the Art and Current Typical Energy Data

If all U.S. petroleum refineries were able to attain onsite SOA energy intensities, it is estimated

that 286 TBtu per year of energy could be saved from the nine processes alone, corresponding to

a 14% energy savings sector-wide. This energy savings estimate is based on adopting available

SOA technologies and practices without accounting for future gains in energy efficiency from

R&D.

Delving deeper, certain petroleum refining processes have relatively large available energy

savings opportunities. In the following Sections, several processes are highlighted and the nature

of the onsite SOA energy savings is discussed. Appendix A1 provides detailed data for each

individual process.

5.2.1.1. Atmospheric Crude Distillation

As the first step of the petroleum refining process, atmospheric distillation plays an important

role and is the largest energy consuming process. About 82 TBtu per year or 29% of the total

SOA energy savings for the sector can be attributed to atmospheric crude distillation. The crude

oil is preheated in a series of heat exchangers in order to preheat and vaporize the crude oil feed

before entering the atmospheric distillation tower. Retrofitting the crude oil preheat train to reach

the highest possible furnace inlet temperature is the main method to reduce the process’s overall

energy use. As mentioned in Section 4.2, higher temperatures can lead to accelerated fouling,

which reduces the effectiveness of the HEN revamp. To maintain good efficiency, an active heat

exchanger cleaning program may have to be undertaken.

Figure 5-2. Current Energy Savings Opportunity by Petroleum Refining Process

Studied (Energy Savings Per Year in TBtu)


5.2.1.2. Fluid Catalytic Cracking

Fluid catalytic cracking (FCC) accounts for 45 TBtu per year or 16% of the total SOA energy

savings for petroleum refining. While the FCC unit can generally be considered as a net energy

producer from a utility standpoint, a large source of the surplus energy comes from the catalyst

regeneration step during which coke is burned off. This energy can be recovered and used for

heating purposes and to provide the endothermic heat of reaction; the catalytic cracking is

typically performed at a high temperature, between 900°F and 1000°F. Power recovery is another

example of an opportunity to improve energy efficiency in FCCs.

5.2.1.3. Catalytic Reforming

Of the total petroleum refining SOA energy savings, 33 TBtu per year or about 12% comes from

catalytic reforming. Catalytic reforming involves four major reactions in the presence of a metal

catalyst, three of which are endothermic (dehydrogenation, dehydrocyclization, and

isomerization) while the other is exothermic (hydrocracking). The temperature of the

hydrocarbon feed stream is typically maintained at about 950°F. The significant amount of

process energy being used as fuel for process heaters is an area to improve energy efficiency.

5.2.1.4. Vacuum Crude Distillation

Of the 286 TBtu per year of available sector-wide SOA energy savings, vacuum crude

distillation accounts for 32 TBtu per year or 11%. Vacuum distillation processes the heavy crude

residue, or bottoms product, from atmospheric crude distillation. The components of this residue

have boiling points above 750°F, resulting in very high temperatures in the equipment and cause

the formation of coke deposits. Operating the distillation tower in a vacuum lowers the

temperature necessary to separate the components and reduce some of the negative effects of the

residue itself on the process.


6. Practical Minimum Energy Consumption for

U.S. Petroleum Refining

Technology innovation is the driving force for economic growth. Across the globe, R&D is

underway that can be used to make petroleum refining products in new ways and improve energy

and feedstock efficiency. Commercialization of these improvements will drive the

competitiveness of U.S. petroleum refining. In this Chapter, the R&D energy savings made

possible through R&D advancements in petroleum refining are estimated. Practical minimum

(PM) is the minimum amount of energy required assuming the deployment of applied R&D

technologies under development worldwide.

6.1. R&D IN THE PETROLEUM REFINING INDUSTRY

As a matter of business strategy, the major U.S. refining companies have been steadily reducing

their R&D budgets for over 40 years, choosing instead to rely on third parties to perform most of

the new technology development. The dissemination of new techniques is then shared through

forums such as technical literature (e.g., publications such as Hydrocarbon Processing, Oil &

Gas Journal, etc.) or industrial seminars or conferences (IPPC 2012).

A 2004 energy efficiency roadmap for California’s petroleum refineries notes that the top

priority areas for R&D in refining in order of significance are: new process technology,

improved process operations, conserving and generating electricity, energy systems/

management, and treatment and sulfur removal (CEC 2004). Because oil refining technology is

fairly standard across the world, these targeted areas of improvement are likely to apply to all

refineries, whether located in the U.S. or elsewhere in the world.

In general, oil and gas companies are very reliant on technology royalties. Research in this sector

is very tightly protected due to the competitive nature of the industry. For the most part, R&D

information is kept confidential until the technology has been developed, piloted, and

successfully licensed. An attempt was made in this study, to solicit research savings estimates

from leading petroleum refining companies. All operating companies declined to cooperate due

to concerns of confidentiality. Only UOP, whose sole business is process development and

technology licensing as a vendor to the oil refining industry, provided a summary of research

savings areas by process unit with corresponding energy savings estimates. Other sources of

savings estimates were consulted and a full list can be found in Section 6.3 below.

It is also worth noting that there are cultural barriers in most oil refining companies towards

adopting innovative new technology, because over the years the industry has become a

commodity business with very thin profit margins. Refinery managers are generally rewarded for

maintaining target production throughput and yield, with minimal downtime. The penalty for

unscheduled production outages (a possible consequence of pursuing of cost savings) far

outweighs any potential benefits. The reality is that despite accounting for 60% of a refinery’s


controllable costs, energy efficiency is usually a secondary priority at best (Kumana &

Associates 2013).

6.2. DESCRIPTION OF A FUTURE REFINERY

Current trends in the petroleum refining industry suggest that the refinery of the future will have

the following characteristics (Kumana & Associates 2013):

Greater capacity; at least 200,000 barrels per day and as much as 600,000 barrels per day

Utilization of multiple feedstock sources; there are only a handful of giant oil fields in the

world capable of supplying all of the crude oil requirements of such large refineries

More complex, with the ability to process heavier (i.e., American Petroleum Institute (API)

gravity less than 20) and higher sulfur crudes; this will require investment in more

hydrotreating and other advanced technologies

Campaign-style operation for the crude distillation/vacuum distillation units (i.e., running

different feedstock blends depending upon crude oil feedstock availability, each one for only

a few days at a time). Some of the larger refineries in India receive crude oil from over 50

different sources. Tank farm operations and optimized blending strategies will become

increasingly important.

Located at or near major sea-ports, pipeline hubs, or major production fields

Co-located with a petrochemicals complex (e.g., 10 large grass-roots refineries built in the

past decade largely fit this profile)

The larger existing refineries will likely continue to expand in production capacity to meet

demand requirements. Energy efficiency will remain a significant concern for refineries as the

price of feedstocks, operations, and compliance increase. Process modifications will be needed

for refineries to operate flexibly and be able to handle feedstocks of varying composition and

quality. Also, refineries will need to be able to produce higher quality and range of products,

including new alternative or more environmentally friendly fuels, from these lower quality

feedstocks in order to keep up with increasing demand.

Refineries will have to comply with any future environmental regulations, whether it is through

emissions allowances or fuel quality requirements. The establishment of different, non-

traditional refineries that take advantage of alternative feedstocks (e.g., biomass, coal, and crude

shale oil) and processes may become more widespread, including biorefineries, coal liquid

refineries, shale oil refineries, and gasification refineries.

Table 6-1 below, adapted from The Refinery of the Future by James G. Speight, provides some

predictions of likely process changes and technology developments that in the nine process units

that are the focus of this bandwidth analysis. Other general cross-cutting development areas

include catalyst innovations and adaptation to new feedstocks.


Table 6-1. Examples of Future Refining Technology and Process Developments*

Process Details

Atmospheric and vacuum crude

distillation

Short-term developments: improved heat recovery and integrating atmospheric and

vacuum distillation

Long-term developments: integration of different distillation columns into one reactor or

alternative process routes that allow for combined conversion/distillation

Alternative processes: membranes or freeze concentration

Catalytic Hydrocracking

Will take central position in integration of refineries and petrochemical plants due to flexibility; development of innovative fixed bed designs that allow better catalyst utilization or online catalyst removal in order to handle increased heavy feedstocks

Coking/visbreaking

Coking: changes to reactor internals and nature of catalysts, but likely to remain as a

commonly used refinery process

Visbreaking: may be increasingly used as pretreatment process

Hydrotreating

Advanced hydrotreating developments: new catalysts, catalytic distillation, and

processing at mild conditions, hydrodesulfurization development will continue for cleaner transportation fuel production

Upgrading approaches: higher activity and more resilient catalysts, replacement of reactor

internals for increased efficiency, adding reactor capacity, specialized process design and hardware

Long-term developments: new desulfurization technologies or evolution of the older

technologies that will reduce the need for hydrogen

* Adapted from Speight 2011

6.3. CALCULATED PRACTICAL MINIMUM ENERGY CONSUMPTION FOR


In this study, PM energy intensity is the estimated minimum amount of energy consumed in a

specific petroleum refining process assuming that the most advanced technologies under research

or development around the globe are deployed.

R&D progress is difficult to predict and potential gains in energy efficiency can depend on

financial investments and market priorities. To estimate PM energy consumption for this

bandwidth analysis, a broad search of R&D activities in the petroleum refining industry was

conducted. This research turned up a very limited range of promising new technologies for

consideration.

The focus of this study’s search was applied research, which was defined as investigating new

technology with the intent of accomplishing a particular objective. Basic research, the search for

unknown facts and principles without regard to commercial objectives, was not considered.

Many of the technologies identified were disqualified from consideration due a lack of data from

which to draw energy savings conclusions.


Appendix A1 presents the onsite PM energy consumption for the nine processes considered in

this bandwidth study in alphabetical order. The PM energy consumption for each process is

calculated by multiplying the estimated PM energy intensity for each process by the process’s

2010 throughput or production volume (the energy intensity and throughput or production data

are also presented in Appendix A1).These values exclude feedstock energy. The lower limit for

onsite PM energy intensity and onsite PM energy consumption are presented in Appendix

A1.The upper limit of the PM range is assumed to be the SOA energy consumption. The PM

energy consumption for each process is expressed as a range because the energy savings impacts

are speculative and based on unproven technologies.

Table 6-2 presents the key sources consulted to identify PM energy intensities in petroleum

refining. Additionally, numerous fact sheets, case studies, reports, and award notifications were

referenced; a more detailed listing of references is provided in Appendix A3 (Table A3 and

References for Table A3).

Table 6-2. Key Published Sources Reviewed to Identify Practical Minimum Energy Intensities for Select

Processes

Reference Abbreviation

Source

ANL 1999 The Potential for Reducing Energy Utilization in the Refining Industry, Argonne National Laboratory, Petrick, M. and Pellegrino, J, 1999.

HP 2008b “Improved crude oil fractionation by distributed distillation”, Hydrocarbon Processing (HP), Haddad, H.N. and Manley, D.B., 2008.

LBNL 2000 Emerging Energy-Efficient Industrial Technologies, Lawrence Berkeley National Laboratory (LBNL), Martin, N., Worrell, E., Ruth, M., Price, L., Elliott, R.N., Shipley, A.M., Thorne, J, 2000.

Szklo & Schaeffer 2007

“Fuel specification, energy consumption, and CO2 emissions in oil refineries,” Energy 32: 1075-1092, Szklo, A. and Schaeffer, R., 2007.

UOP 2013 Industrial contact estimates

Numerous fact sheets, case studies, reports, and award notifications were referenced. Details of all of the practical minimum sources consulted can be found in Appendix A3.

Appendix A3 presents details on the R&D technologies that were selected and used to estimate

the PM energy intensities. Energy savings from R&D advancements were directly estimated for

the nine processes. In Appendix A3, technologies are aligned with the most representative

process. Some of the technologies have applicability to more than one process (e.g., both

atmospheric and vacuum distillation).

Analysis of the range of energy savings offered by groups of technologies is complicated in

that the savings offered by multiple technologies may or may not be additive. Each technology

contributes discrete or compounding savings that increase the ultimate savings of the group and

some energy savings may be duplicative. As a result, all values are presented as sourced from the

literature and energy savings were not aggregated for multiple technologies. A separate study of

the individual technologies would be necessary to verify and validate the savings estimates and

interrelationships between the technologies. If more than one technology was considered for a


particular process, the technology that resulted in the lowest energy intensity was conservatively

selected for the PM energy intensity.

R&D in some process areas is more broadly applicable, such as utility/power generation

improvements and crosscutting technologies. Cross-cutting technologies applied during the PM

analysis included new high-temperature, low-cost ceramic media for natural gas combustion

burners, the application of modeling and process analysis, and advanced energy and water

recovery technology from low-grade waste heat. The estimated energy savings from utility and

crosscutting improvements were assumed to be applicable to all nine processes studied. To

calculate PM energy consumption, the CT energy intensity and TM energy intensity were

multiplied by the combined estimated savings for utility and crosscutting improvements (27%-

39%) and subtracted from the CT energy consumption.

In Appendix A3, the range of technologies considered offer a corresponding range of estimated

energy savings. Brief descriptions of the technologies are followed by reported savings in terms

of dollars, Btu, and percent savings. The technology developers' estimated savings were taken at

face value and adjusted to represent the overall average energy savings potential.

For each technology, Appendix A3 presents a brief explanation of the energy savings and a

summary of adjustments necessary to determine the overall average energy savings potential and

PM energy intensity. Research savings are speculative in nature. The energy savings will vary

depending on the source; they can be reported in terms of primary energy savings, refinery-wide

energy savings, process energy savings, or energy-type savings. In each case, the reported energy

savings were adjusted to determine PM energy intensity.

6.3.1. Weighting of Technologies

The technologies described in Appendix A3 can be weighted differently depending on the

audience. Plant managers may primarily be interested in productivity and quality implications;

business managers may primarily be interested in relative cost and payback; technology investors

may primarily be interested in market impact, technology readiness, and development risk

factors; and government regulators may primarily be interested in environmental impacts. Each

factor plays heavily into R&D investment considerations.

Appendix A4 (Table A4) considers how to weigh these various perspectives. Six technology

weighting factors were considered for each technology:

A Technology Readiness

B Market Impact

C Relative Cost and Savings Payback

D Technical Risk

E Productivity/Product Quality Gain

F Environmental Impacts


Appendix A4 (Table A4) presents the PM technology weighting factors that could be applied to

the technologies for specific processes (as identified in Appendix A3). Best engineering

judgment was employed to rate each of the technologies with these weighting factors. A score of

High, Medium, or Low was assigned to each factor along with a brief explanation for the score.

The parameters referenced in scoring are detailed in Appendix A4 (Table A4). An overall

importance rating for the technology was determined based on the weighting factor scores. Each

weighting factor is assigned a DOE importance level of “1.”This importance level can be altered;

for example, if Technology Readiness and Market Impact carry higher importance, the

importance level for these factors can be changed to “2” or “3” and the resulting Overall

Importance Rating would change accordingly.

The weighting factors presented in Appendix A4 can be used for further study of the R&D

technologies identified in Appendix A3. The weighting factor study was part of the analysis of

the R&D technologies, and serves as a guide for prioritizing the technologies. However, the

weighting factors were not utilized to estimate onsite PM energy intensity or consumption.

6.4. PRACTICAL MINIMUM ENERGY CONSUMPTION BY PROCESS AND

SECTOR-WIDE

Table 6-3 presents the onsite PM energy consumption for the nine processes studied and

petroleum refining sector-wide. The onsite PM energy savings is the difference between CT

energy consumption and PM energy consumption. PM energy savings is equivalent to the sum of

current and R&D opportunity energy savings.

In Table 6-3, data is extrapolated to estimate the total PM subsector opportunity. PM energy

consumption for the individual processes studied is summed and the data is extrapolated to

estimate sector total. PM subsector energy savings is also expressed as a percent in Table 6-3.

This is also shown in Figure 6-1. It is useful to consider both TBtu energy savings and energy

savings percent when comparing energy savings opportunity. Both are good measures of

opportunity; however, the conclusions are not always the same.


Table 6-3. Onsite Practical Minimum Energy Consumption, Energy Savings, and Energy Savings Percent for

the Processes Studied and Sector-Wide

Process

Onsite CT

Energy

Consumption

(TBtu/year)

Onsite PM

Energy

Consumption

(TBtu/year)

PM Energy

Savings †

(CT-PM)

(TBtu/year)

PM Energy

Savings

Percent

(CT-PM)/

(CT-TM)*

Alkylation 90 56-80 10-34 9-30%

Atmospheric Crude Distillation 604 314-522 82-290 20-70%

Catalytic Hydrocracking 85 57-74 11-27 9-22%

Catalytic Reforming 279 188-246 33-90 17-46%

Coking/ Visbreaking 114 66-101 12-48 11-42%

Fluid Catalytic Cracking 334 243-289 45-92 17-35%

Hydrotreating 390 252-333 57-138 9-21%

Isomerization 44 25-39 5-19 11-43%

Vacuum Crude Distillation 223 135-192 32-88 23-64%

Total for Processes Studied 2,163 1,336-1,877 286-826 14-40%

All Other Processes 1,013 626-879 134-387 14-40%

Total for Petroleum Refining

Sector-wide 3,176 1,963-2,756** 420-1,213 14-40%

Current typical (CT), Practical minimum (PM), Thermodynamic Minimum (TM)

† PM energy savings is the Current Opportunity plus the R&D Opportunity.

* Calculated using TM from Table 7-2 as the minimum energy of production. This accounts for the energy necessary to perform the

process. Potential opportunity reflects the difference between CT and TM energy consumption. Calculation: (CT- PM)/(CT- TM).

** The sector-wide PM energy consumption was an extrapolated value, calculated by dividing the total onsite SOA energy

consumption for the processes studied by the overall percent coverage from Table 4-3 (68%).

Figure 6-1 presents the current opportunity and the R&D opportunity for each process; the

current opportunity is the difference between CT energy consumption and SOA energy

consumption (shown in blue) and the R&D opportunity is the difference between the SOA

energy consumption and the PM energy consumption (shown in green). In Figure 6-1, the

percent savings is the percent of the overall energy consumption bandwidth where TM is the

lower baseline. For the processes studied, the greatest current opportunity in terms of percent

savings is vacuum crude distillation at 23% energy savings; the greatest R&D opportunity is

atmospheric crude distillation at 70% energy savings. In Figure 6-2, the current and R&D

savings opportunity is shown in terms of TBtu per year savings. The pie chart in Figure 6-2

captures the blue and green portions of the bar chart shown in Figure 6-1, each in a separate pie

chart. The greatest current opportunity in terms of TBtu savings is atmospheric crude distillation

at 82 TBtu per year savings; the greatest R&D opportunity in terms of TBtu savings is also

atmospheric crude distillation at 290 TBtu per year savings. In terms of both percent energy

savings and TBtu/year savings, this process shows the greatest overall opportunity.


To extrapolate the data for all other processes that is shown in Table 6-3 and Figure 6-2, the PM

energy consumption of each individual process studied is summed, and the sum is divided by the

percent coverage for the entire sector (68%, see Table 4-3). The percent coverage of processes

studied compared to the total CT energy consumption of the sector is shown in the last column of

Table 4-3. Percent coverage is the ratio of the sum of all the CT energy consumption for the

individual processes studied to the CT energy consumption for the sector provided by MECS

(see Table 4-3). The PM energy consumption for the remainder of the sector (i.e., all processes

that are not included in the nine processes studied) was calculated by subtracting the total for the

processes studied from the sector-wide total. These additional processes are together referred to

as All Other Processes in Table 6-3.

Table 6-3 also presents the PM energy savings percent. To calculate the onsite PM energy

savings percent, the thermodynamic minimum (TM) energy consumption serves as the baseline

for estimating percent energy savings, not zero. The energy savings percent is the percent of

energy saved with PM energy consumption (i.e., the deployment of R&D technologies under

development worldwide) compared to CT energy consumption, considering that the TM energy

consumption may not be zero (i.e., the TM energy consumption may be negative). As will be

explained in Chapter 7, in some cases, the TM reaction energy is a negative value. When

comparing energy savings percent from one process to another (or one subsector to another), the

absolute savings is the best measure of comparison. The equation for calculating onsite PM

energy savings percent is:

𝑃𝑀 𝑆𝑎𝑣𝑖𝑛𝑔𝑠 % = 𝐶𝑇 − 𝑃𝑀



Figure 6-1. Current and R&D Opportunity Energy Savings Bandwidths for the Petroleum Refining

Processes Studied (with Percent of Overall Energy Consumption Bandwidth)


The PM energy savings opportunity is different than SOA energy savings opportunity in that the

scope of the R&D technologies contributing energy savings can essentially be boundless. Putting

aside obvious financial, timing, and resource limitations, the process improvements and

increased energy efficiency that can be gained through unproven technology is speculative. For

this reason, a range is used to represent the potential onsite PM energy consumption, PM energy

savings, and PM energy savings percent in Table 6-3. The upper limit of the PM energy

consumption range is assumed to be equal to the SOA energy consumption. The lower limit of

the PM energy consumption range was estimated using the method explained in Section 6.2. The

lower limit is shown as a dashed line with color fading in the summary figures that present

Figure 6-2. Current and R&D Energy Savings Opportunities by Petroleum Refining

Process Studied (Energy Savings Per Year in TBtu)


subsector and sector-wide data. This is done because the PM is speculative and depends on

unproven R&D technologies; furthermore, the potential energy savings opportunity could be

bigger if additional unproven technologies were considered.

6.4.1. Atmospheric and Vacuum Crude Distillation Practical Minimum

Energy Savings

While energy intensity and overall consumption will vary for different refineries, depending

upon feedstock composition, capacity, and other factors, there are alternative distillation

equipment, processes, and techniques that are being given extensive R&D consideration,

especially when compared to the other seven process units. There are numerous ways to reduce

the energy consumption of atmospheric and vacuum distillation, and together these two process

units offer significant R&D opportunity savings. Thermal cracking is one example of a process

that could conceivably replace crude distillation, wherein thermally intensive distillation is

replaced by less thermally intensive- cracking. Compared to atmospheric and crude distillation,

thermal cracking would consume about 25% less energy (ANL 1999).

As mentioned in Table 6-1, heat recovery and integrating atmospheric and vacuum crude

distillation are both examples of developments that can help reduce the energy consumption of

these process units. In the example of progressive distillation, two separation processes –gasoline

fractionation and a naphtha stabilizer− are integrated, resulting in savings of about 30% in

energy. Dividing wall columns are another example of integrating atmospheric and vacuum

crude distillation using a vertical partition. Although both of these new types of distillation have

been applied to select refineries in a limited fashion, more development is needed to make the

technology commercially acceptable. For heat recovery, self-heat recuperation, or where whole-

process heat is re-circulated so that no heat needs to be added, is one example of how the energy

efficiency of atmospheric distillation could be included (Kansha et al. 2012).


7. Thermodynamic Minimum Energy

Consumption for U.S. Petroleum Refining

Real world petroleum refining does not occur under theoretically ideal conditions; however,

understanding the theoretical minimal amount of energy required to manufacture a petroleum

refining product can provide a more complete understanding of opportunities for energy savings.

This baseline can be used to establish more realistic projections of what future R&D energy

savings can be achieved. This Chapter presents the thermodynamic minimum (TM) energy

consumption required for the processes studied and for the entire sector.

7.1. THERMODYNAMIC MINIMUM ENERGY

TM energy consumption is the calculated minimum amount of energy theoretically needed to

complete a petroleum refining process, assuming ideal conditions that are typically unachievable

in real-world applications; in some cases, it is less than zero. TM energy consumption assumes

all the energy is used productively and there are no energy losses. It is based on the Gibbs Free

Energy (ΔG) equation under ideal conditions for a process. Some petroleum refining processes

are net producers of energy (i.e., exothermic processes); this created energy was considered in

this analysis.

7.2. CALCULATED THERMODYNAMIC MINIMUM ENERGY

CONSUMPTION FOR INDIVIDUAL PROCESSES

Appendix A1 presents the onsite TM energy consumption for the nine processes considered in

this bandwidth study in alphabetical order. For a given process, the TM energy intensity is

multiplied by the annual U.S. production or throughput to determine the total onsite TM energy

consumption (the energy intensity and production/throughput data are also presented in

Appendix A1). Table 7-1 presents the references for the TM energy intensity values and the

applicable process. Appendix A2 also provides the references for the TM energy intensity data

for each individual process.

Table 7-1. Published Sources Reviewed to Identify Thermodynamic Minimum Energy Intensities for the

Processes Studied

Source Process

DOE 2006 Catalytic reforming, alkylation, and fluid catalytic cracking

Internal calculations based on change in Gibbs free

energy at ideal conditions (see Appendix A5 for

details)

Atmospheric and vacuum distillation, hydrotreating, catalytic hydrocracking, isomerization, and coking/visbreaking

Petroleum refining can at times result in net energy gain through exothermic processes; this is

the case for a few processes studied (e.g., hydrotreating). For exothermic petroleum refining

processes, a zero baseline would result in negative percent savings, a physical impossibility. TM


energy consumption was instead referenced as the baseline (or minimum amount of energy)

when calculating the absolute energy savings potential. The equations used to determine the

absolute energy savings for SOA and PM are as follows:

𝑆𝑂𝐴 𝑆𝑎𝑣𝑖𝑛𝑔𝑠 % = 𝐶𝑇 − 𝑆𝑂𝐴


𝑃𝑀 𝑆𝑎𝑣𝑖𝑛𝑔𝑠 % = 𝐶𝑇 − 𝑃𝑀


For processes requiring an energy intensive transformation (e.g., atmospheric crude distillation),

this percent energy savings approach results more realistic and comparable energy savings

estimates. Using zero as the baseline (or minimum amount of energy) would exaggerate the total

bandwidth to which SOA energy savings and PM energy savings are compared to determine the

energy savings percent. When TM energy consumption is referenced as the baseline, SOA

energy savings and PM energy savings are relatively more comparable, resulting in more

accurate energy savings percentages.

7.2.1. Calculated Thermodynamic Minimum Energy Intensities

As noted in Table 7-1, the TM energy intensity for six of the nine processes studied was

calculated based on the change in Gibbs free energy (ΔG) at ideal conditions. Full assumptions,

explanations, and other information on how these TM energy intensities were calculated can be

found in Appendix A5.

For atmospheric and vacuum crude distillation, the overall head balance can be described by:

𝑇ℎ𝑒𝑟𝑚𝑜𝑑𝑦𝑛𝑎𝑚𝑖𝑐 𝑀𝑖𝑛𝑖𝑚𝑢𝑚 𝐸𝑛𝑒𝑟𝑔𝑦 𝐼𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦 = 𝐸𝑛𝑒𝑟𝑔𝑦 𝑂𝑢𝑡 − 𝐸𝑛𝑒𝑟𝑔𝑦 𝐼𝑛

The TM energy intensity was calculated as the change in Gibbs free energy, ΔGdist, which was

equal to 69,313 Btu per barrel. Catalytic hydrotreating is the process by which hydrogen is added

to reduce double bonds and to remove sulfur, nitrogen, oxygen and halogen impurities from the

feedstock. To calculate the change in energy, ΔGrxn was calculated for each reaction at the

temperature of the reactor, and was summed to determine the overall ΔG or TM energy intensity

of -53,000 Btu per barrel. Hydrocracking is an exothermic process, which replaces a carbon-

carbon single bond and a hydrogen-hydrogen single bond with two carbon-hydrogen bonds. The

TM energy intensity for hydrocracking was calculated to be equal to -71,900 Btu per barrel,

although this number varies directly with the amount of hydrogen needed to hydrocrack the

incoming feed stream.

The isomerization process is exoergonic; based on the assumptions of the inputs and output

streams (found in Appendix A5), the TM energy intensity was calculated as -4.91 Btu per barrel,

though this number is strongly dependent upon which branched alkanes are favored by the

catalyst and operating conditions. Finally for coking and visbreaking, the TM energy intensity


was assumed to be equal to the TM energy intensity for delayed coking (calculated as 350 Btu

per barrel) because it consumes a majority of the energy for the coking/visbreaking process.

7.3. THERMODYNAMIC MINIMUM ENERGY CONSUMPTION BY

PROCESS AND SECTOR-WIDE

The minimum baseline of energy consumption for a petroleum refining process is its TM energy

consumption. If all the 2010 level of petroleum refining production occurred at TM energy

intensity, there would be 100% savings. The percentage of energy savings is determined by

calculating the absolute decrease in energy consumption and dividing it by the total possible

savings (CT energy consumption-TM energy consumption).

Table 7-2 provides the TM energy consumption for the nine processes studied (excluding

feedstock energy). In theory, if heat generating processes could be carefully coupled with heat

consuming processes, this could greatly offset the energy usage in petroleum refining overall. It

is an imperative to keep in mind that ideal conditions are largely unrealistic goals in practice and

these values serve only as a guide to estimating energy savings opportunities.

Table 7-2 also presents the extrapolated TM energy consumption for the entire sector. The

extrapolation for sector-wide TM energy consumption is done with the same methodology as for

SOA energy consumption and PM energy consumption (as explained in Section 5.2 and 6.4).

The TM energy consumption was used to calculate the current and R&D energy savings

percentages (not zero).


Table 7-2. Thermodynamic Minimum Energy Consumption by Process and

Sector-Wide for the Nine Processes Studied and Extrapolated to Sector Total

Process

Onsite TM

Energy Consumption

(TBtu/year)

Alkylation -21

Atmospheric Crude Distillation 188

Catalytic Hydrocracking -38

Catalytic Reforming 83

Coking/ Visbreaking 0

Fluid Catalytic Cracking 73

Hydrotreating -256

Isomerization 0

Vacuum Crude Distillation 86

Total for Processes Studied 115

All Other Processes 54

Total for Petroleum Refining Sector-wide 169*

Thermodynamic minimum (TM)

† Estimates for the entire sector were extrapolated by dividing the onsite TM energy consumption for the processes studied by the overall percent coverage of 68% (see Table 4-3).


8. U.S. Petroleum Refining Energy Bandwidth

Summary

This Chapter presents the energy savings bandwidths for the petroleum refining processes

studied and sector-wide based on the analysis and data presented in the previous Chapters and

the Appendices. Data is presented for the nine processes studied and extrapolated to estimate the

energy savings potential for all of U.S. petroleum refining.

8.1. PETROLEUM REFINING BANDWIDTH PROFILE

Table 8-1 presents the current opportunity and R&D opportunity energy savings for the nine

processes studied and extrapolated to estimate the sector total. The process totals are summed to

provide a sector-wide estimate. The energy savings data was extrapolated to account for all other

processes not included in the nine processes studied, as explained in Section 5.2 (SOA) and 6.4

(PM). Each row in Table 8-1 shows the opportunity bandwidth for a specific petroleum refining

process and sector-wide.

As shown in Figure 8-1, four hypothetical opportunity bandwidths for energy savings are

estimated (as defined in Chapter 1). To complete the nine processes studied, the analysis shows

the following:

Current Opportunity – 286 TBtu per year of energy savings could be obtained if state of

the art technologies and practices are deployed.

R&D Opportunity – 540 TBtu per year of additional energy savings could be attained in

the future if applied R&D technologies under development worldwide are deployed (i.e.,

reaching the practical minimum).

To complete all of the U.S. petroleum refining sector processes (based on extrapolated data), the

analysis shows the following:

Current Opportunity – 420 TBtu per year of energy savings could be obtained if state of

the art technologies and practices are deployed.

R&D Opportunity – 793 TBtu per year of additional energy savings could be attained in

the future if applied R&D technologies under development worldwide are deployed (i.e.,

reaching the practical minimum).

Figure 8-1 also shows the estimated current and R&D energy savings opportunities for

individual petroleum refining processes. The area between R&D opportunity and impractical is

shown as a dashed line with color fading because the PM energy savings impacts are speculative

and based on unproven technologies.


Table 8-1. Current Opportunity and R&D Opportunity Energy Savings for the Nine

Processes Studied and Extrapolated to Sector-Wide Total

Process

Current

Opportunity

(CT-SOA)

(TBtu/year)

R&D

Opportunity

(SOA-PM)

(TBtu/year)

Alkylation 10 24

Atmospheric Crude Distillation 82 208

Catalytic Hydrocracking 11 17

Catalytic Reforming 33 58

Coking/Visbreaking 12 36

Fluid Catalytic Cracking 45 47

Hydrotreating 57 81

Isomerization 5 14

Vacuum Crude Distillation 32 56

Total for Processes Studied 286 540

All Other Processes 134 253

Total for Petroleum Refining Sector-wide 420 793

Current typical (CT), state of the art (SOA), practical minimum (PM)

From the processes studied, the greatest current and R&D energy savings opportunity for

petroleum refining comes from upgrading production methods in atmospheric crude distillation

and hydrotreating.

The impractical bandwidth represents the energy savings potential that would require

fundamental changes in petroleum refining. It is the difference between PM energy consumption

and TM energy consumption. The term impractical is used because the significant research

investment required based on today’s knowledge would no longer be practical because of the

thermodynamic limitations. The TM energy consumption is based on ideal conditions that are

typically unattainable in commercial applications. It was used as the baseline for calculating the

energy savings potentials (not zero) to provide more accurate targets of energy savings

opportunities.


Figure 8-2 shows the bandwidth summaries for the petroleum refining processes presented in

order of highest current plus R&D energy savings opportunity. Atmospheric crude distillation is

the largest energy consuming process in petroleum refining. If the lower limit of PM energy

consumption could be reached, this would save about 290 TBtu/year compared to CT, amounting

to 9% of CT energy consumption for the entire petroleum refining sector. Other processes, such

as alkylation, catalytic hydrocracking, and isomerization, have a much smaller difference

between CT energy consumption and the PM energy consumption. Figure 8-2 shows the relative

size of the current and R&D opportunity energy savings potential for each process.

Figure 8-1. Current and R&D Energy Savings Opportunities in U.S. Petroleum Refining for the Processes Studied

and for Sector-Wide based on Extrapolated Data


Figure 8-2. Current and R&D Opportunity Energy Savings Bandwidths for the Processes Studied (with

Percent of Overall Energy Consumption Bandwidth)


9. References

AIChE 2000

Refining Overview- Petroleum Process & Products (CD-ROM). Self, F.,

Ekholm, E., and Bowers, K. American Institute of Chemical Engineers (AIChE).

2000.

Al-Mutairi

2010

“Optimal Design of Heat Exchanger Network in Oil Refineries.” Al-Mutairi,

E.M. Chemical Engineering Transactions 21(2010): 955-960. 2010.

ANL 1999

The Potential for Reducing Energy Utilization in the Refining Industry. Argonne

National Laboratory (ANL). Petrick, M. and Pellegrino, J. 1999.

www.ipd.anl.gov/anlpubs/1999/10/34061.pdf

Bulasara et

al. 2009

“Revamp study of crude distillation unit heat exchanger network: Energy

integration potential of delayed coking unit free hot streams.” Bulasara, V.K.,

Uppaluri, R., Ghoshal, A.K. Applied Thermal Engineering 29 (2009): 2271-

2279. 2009.

CEC 2004

Energy Efficiency Roadmap for Petroleum Refineries in California. Prepared by

Energetics, Inc. for the California Energy Commission (CEC). 2004.

http://energy.gov/sites/prod/files/2013/11/f4/refining_roadmap.pdf

Chang et al.

2012

“Characterization, Physical and Thermodynamic Properties of Oil Fractions.” In

Refinery Engineering: Integrated Process Modeling and Optimization, 1st ed.

Chang, A.F., Pashikanti, K., and Y.A. Liu. 2012. Wiley-VCH Verlag GmbH &

Co. www.wiley-vch.de/books/sample/3527333576_c01.pdf

ChemEd DL

n.d. “Bond Enthalpies.” Chemical Education Digital Library (ChemEd DL). n.d.

Colorado

School of

Mines n.d.

“Delayed Coking: Chapter 5.” Presentation. Colorado School of Mines. n.d.

DOE 1999

“Effective Fouling Minimization Increases the Efficiency and Productivity of

Refineries.” U.S. Department of Energy, Office of Industrial Technologies.

January 1999. www.energy.gov/sites/prod/files/2014/05/f16/foulmitpet.pdf

DOE 2003

“Gasoline Biodesulfurization.” U.S. Department of Energy, Industrial

Technologies Program. May 2003.

http://energy.gov/sites/prod/files/2014/05/f16/gasbiopet_final.pdf

DOE 2006

Energy Bandwidth for Petroleum Refining Processes. Prepared by Energetics,

Inc. for U.S. Department of Energy (DOE), Office of Industrial Technologies

(OIT). 2006. http://energy.gov/sites/prod/files/2013/11/f4/bandwidth.pdf


DOE 2007

Energy and Environmental Profile of the U.S. Petroleum Refining Industry.

Prepared by Energetics, Inc. for U.S. Department of Energy (DOE), Office of

Industrial Technologies (OIT). 2007.

http://energy.gov/sites/prod/files/2013/11/f4/profile.pdf

DOE 2011

Grand Challenge Portfolio: Driving Innovations in Industrial Energy Efficiency.

U.S. Department of Energy, Industrial Technologies Program. 2011.

www1.eere.energy.gov/manufacturing/pdfs/grand_challenges_portfolio.pdf

DOE 2013

“Energy Savings Assessments (ESA).” Conducted by the U.S. Department of

Energy (DOE) and Oak Ridge National Laboratory (ORNL) between 2006 and

2011.

DOE 2014

“Manufacturing Energy and Carbon Footprints”, U.S. Department of Energy

(DOE), Advanced Manufacturing Office (AMO) (formerly Industrial

Technologies Program, or ITP), 2014.

http://energy.gov/eere/amo/manufacturing-energy-and-carbon-footprints-2010-

mecs

Downing

2011

“Increase Your Boiler Pressure to Decrease You Electric Bill: The True Cost of

CHP.” Proceedings of the Industrial Energy Technology Conference. Downing,

A. May 2011. http://repository.tamu.edu/bitstream/handle/1969.1/94796/ESL-

IE-11-05-19.pdf?sequence=1

EIA 2011

“Table 5.9 Refinery Capacity and Utilization, Selected Years, 1949-2011.”

Annual Energy Review 2011. U.S. Energy Information Administration (EIA).

2011. www.eia.gov/totalenergy/data/annual/pdf/sec5_23.pdf

EIA 2012a

“Downstream Charge Capacity of Operable Petroleum Refineries.” U.S. Energy

Information Administration (EIA). 2012.

www.eia.gov/dnav/pet/pet_pnp_capchg_dcu_nus_a.htm

EIA 2012b

“Production Capacity of Operable Petroleum Refineries.” U.S. Energy


www.eia.gov/dnav/pet/pet_pnp_capprod_dcu_nus_a.htm

EIA 2012c

“Table 1: Number and Capacity of Operable Petroleum Refineries by PAD

District and State as of January 1, 2012.” Refinery Capacity 2012. U.S. Energy


www.eia.gov/petroleum/refinerycapacity/table1.pdf

EIA 2012d

“Capacity of Operable Petroleum Refineries by State and Individual Refinery as

of January 1, 2012.” Annual Refinery Report Form EIA-820. U.S. Energy


http://www.eia.gov/petroleum/refinerycapacity/archive/2012/refcap2012.cfm


EIA 2012e

“What are the major sources and users of energy in the United States?” U.S.

Energy Information Administration (EIA). May 2012.

www.eia.gov/energy_in_brief/article/major_energy_sources_and_users.cfm

EIA 2013a

Manufacturing Energy and Consumption Survey 2010. U.S. Energy Information

Administration (EIA). 2013.

www.eia.gov/consumption/manufacturing/data/2010/

EIA 2013b

“Refinery Net Production.” U.S. Energy Information Administration (EIA).

2013. www.eia.gov/dnav/pet/pet_pnp_refp2_dc_nus_mbbl_a.htm

EIA 2013c “Refinery Utilization and Capacity.” U.S. Energy Information Administration

(EIA). 2013. www.eia.gov/dnav/pet/pet_pnp_unc_dcu_nus_a.htm

EIA 2013d “U.S. States: State Profiles and Energy Estimates.” U.S. Energy Information

Administration (EIA). 2013. www.eia.gov/state/

EIA 2013e “International Energy Statistics.” U.S. Energy Information Administration

(EIA). 2013. www.eia.gov/cfapps/ipdbproject/IEDIndex3.cfm

EIA 2013f

“Fuel Consumed at Refineries” U.S. Energy Information Administration (EIA).

2013. www.eia.gov/dnav/pet/pet_pnp_capfuel_dcu_nus_a.htm

ExxonMobil

2011

2011 Corporate Citizenship Report. ExxonMobil. 2011.

www.exxonmobil.com/Corporate/Files/news_pub_ccr2011.pdf

Gary et al.

2007

Petroleum Refining: Technology and Economics, 5th

edition. Gary, J.H.,

Handwerk, G.E., and Kaiser, M.J. New York, NY: Marcel Dekker, Inc. 2007.

GTI 2011

Final Technical Report: Advanced Energy and Water Recovery Technology from

Low Grade Waste Heat. Gas Technology Institute. 2011.

www.osti.gov/bridge/servlets/purl/1031483/1031483.pdf

HP 2006 2006 Refining Processes Handbook. Hydrocarbon Processing (HP). 2006.

HP 2008a 2008 Refining Processes Handbook. Hydrocarbon Processing (HP). 2008.

HP 2008b “Improved crude oil fractionation by distributed distillation”, Hydrocarbon

Processing (HP), Haddad, H.N. and Manley, D.B., 2008.

HP 2011 2011 Refining Processes Handbook. Hydrocarbon Processing (HP). 2011.

IPPC 2012

Best Available Techniques (BAT) Reference Document for the Refining of

mineral oil and gas. Final Draft. European Commission. Integrated Pollution

Prevention and Control (IPPC) Bureau. 2013.

http://eippcb.jrc.ec.europa.eu/reference/BREF/FD_REF_July_2013online.pdf


Kansha et al.

2012

“Application of the self-heat recuperation technology to crude oil distillation.”

Kansha, Y., Kishimoto, A., and Tsutsumi, A. Applied Thermal Engineering

43(2012): 153-157. 2012.

Kumana &

Associates

2013

J D Kumana, Kumana & Associates, Houston, Texas. Personal communication.

2013.

LBNL 2000

Emerging Energy-Efficient Industrial Technologies. Lawrence Berkeley

National Laboratory (LBNL). Martin, N., Worrell, E., Ruth, M., Price, L.,

Elliott, R.N., Shipley, A.M., Thorne, J. 2000.

https://escholarship.org/uc/item/5jr2m969#page-2

LBNL 2005a

Energy Efficiency Improvement and Cost Saving Opportunities for Petroleum

Refineries. Worrell, E. and Galitsky, C. Lawrence Berkeley National Laboratory

(LBNL). 2005.

LBNL 2005b

Assessment of Energy Use and Energy Savings Potential in Selected Industrial

Sectors in India. Sathaye, J., Price, L., de la Rue du Can, S., and Fridley, D.

Lawrence Berkeley National Laboratory (LBNL). 2005.

http://eetd.lbl.gov/node/49899

McKinsey

2009

Pathways to a Low-Carbon Economy. McKinsey & Company. 2009.

www.mckinsey.com/client_service/sustainability/latest_thinking/pathways_to_a

_low_carbon_economy

Nelson 1958 Petroleum Refinery Engineering, 4

th ed. Nelson, W.L. New York: McGraw Hill.

1958.

NRC 2004

Pinch Analysis: For the Efficient Use of Energy, Water & Hydrogen. Oil

Refining Industry: Energy Recovery at a Fluid Catalytic Cracking (FCC) Unit.

Natural Resources Canada (NRC). 2004.

NRC 2005

Pinch Analysis: For the Efficient Use of Energy, Water & Hydrogen. Oil

Refining Industry: Energy Recovery at a Delayed Coker Unit. Natural Resources

Canada (NRC). 2005.

Riazi 2007 Characterization and Properties of Petroleum Fractions. Riazi, M.R. ASTM

Stock Number: MNL 50. USA: ASTM International. 2007.

SBA 2012

“Table of Small Business Size Standards Matched to North American Industry

Classification System.” U.S. Small Business Administration (SBA). 2012.

www.sba.gov/sites/default/files/files/Size_Standards_Table.pdf

SEP 2013 “SEP Certified Facilities.” Superior Energy Performance (SEP). 2013.

http://energy.gov/eere/amo/certified-facilities


Speight 2011 The Refinery of the Future. Speight, J. G. United Kingdom: Gulf Professional

Publishing. 2011.

Szklo &

Schaeffer

2007

“Fuel specification, energy consumption, and CO2 emissions in oil refineries.”

Energy 32: 1075-1092. Szklo, A. and Schaeffer, R. 2007.

Tarighalesla

mi et al. 2011

“An Exergy Analysis on a Crude Oil Atmospheric Distillation Column.”

Tarighaleslami, A.H., Omidkhah, M.R., and Sinaki, S.Y. Chemical Engineering

Transactions 25(2011):117-122. 2011.

UOP 2009

UOP FCC Design Advancements to Reduce Energy Consumption and CO2

Emissions. Wolschlag, L.M., Couch, K.A., Zhu, F., and Alves, J. UOP LLC.

2009. www.honeywell-uop.cn/wp-content/uploads/2011/02/UOP-FCC-Energy-

Optimization-tech-paper1.pdf.

UOP 2013 UOP, a Honeywell Company. Personal communications. 2013.

USCB 2012a “North American Industry Classification System.” United States Census Bureau

(USCB). 2012. www.census.gov/eos/www/naics/index.html

USCB 2012b “Annual Survey of Manufacturers.” United States Census Bureau (USCB). 2012.

www.census.gov/manufacturing/asm/index.html

Varga et al.

2010

“Process Simulation for Improve energy efficiency, maximize asset utilization

and increase in feed flexibility in a crude oil refinery.” Varga, Z., Rabi, I., Stocz,

K.K. Chemical Engineering Transactions 21(2010): 1453-1458.

Zhang et al.

2012

“Sustaining high energy efficiency in existing processes with advanced process

integration technology.” Zhang, N., Smith, R., Bulatov, I., and Klemes, J.J.

Applied Energy. 2012.


Appendix A1: Master Petroleum Refining Table

Table Aa. U.S. Production Volume of Nine Select Petroleum Refining Processes in 2010 with Energy Intensity Estimates and Calculated Onsite

Energy Consumption for the Four Bandwidth Measures (Excludes Feedstock Energy)

Process

2010 Throughput

or Production*

(million bbl)

Energy Intensity

(Btu/bbl)

Calculated Onsite Energy Consumption

(TBtu/year)

CT SOA PM

Lower Limit

TM**

CT SOA PM

Lower Limit

TM

Alkylation 365 246,700 219,400 154,400 -58,000 90.2 80.2 56.4 -21.2

Atmospheric Crude Distillation 5,540 109,100 94,300 56,700 34,000 604.4 522.4 314.3 188.3

Catalytic Hydrocracking 532 158,900 139,000 107,400 -71,900 84.6 74.0 57.2 -38.3

Catalytic Reforming 1,055 263,900 233,000 178,400 79,000 278.5 245.9 188.3 83.4

Coking/Visbreaking 770 147,700 131,700 85,100 350 113.8 101.4 65.6 0.3

Fluid Catalytic Cracking 1,827 182,800 158,300 132,700 40,000 334.0 289.3 242.5 73.1

Hydrotreating 4,829 80,800 68,900 52,200 -53,000 390.2 332.7 252.0 -255.9

Isomerization 203 216,000 192,500 122,300 -5 43.9 39.1 24.8 0.0

Vacuum Crude Distillation 2,504 89,100 76,500 54,000 34,200 223.1 191.5 135.2 85.6

* Values for alkylation and isomerization are production; all other process values are throughput

** Based on previous bandwidth and author calculations

The four bandwidth measures are current typical (CT), state of the art (SOA), practical minimum (PM), and thermodynamic minimum (TM).


Appendix A2: References for U.S. Throughput/Production Data of

the 9 Processes Studied and Energy Intensity Data Used to

Calculate the Current Typical, State of the Art, and Thermodynamic

Minimum Energy Consumption Bands

Table A2. References for U.S. Throughput Data of the Nine Petroleum Refining Processes Studied and Energy Intensity Data Used to Calculate

the Current Typical, State of the Art, and Thermodynamic Minimum Energy Consumption Bands

Process Throughput/ Production

Reference(s)

Throughput/ Production Value

Description

CT Energy Intensity

Reference(s)

SOA Energy Intensity Reference(s)

TM Energy Intensity Reference

Alkylation EIA 2012b; EIA 2013c

Based on alkylate production capacity and 2010 capacity factor of 86.4%

DOE 2007 LBNL 2005a; SEP 2013; DOE 2013; ExxonMobil 2013; McKinsey 2009

DOE 2006

Atmospheric Crude Distillation

EIA 2013c Gross input to atmospheric distillation units

DOE 2007

LBNL 2005a; SEP 2013; DOE 2013; ExxonMobil 2013; McKinsey 2009; LBNL 2005b; Tarighaleslami et al. 2011; Zhang et al. 2012; Varga et al. 2010; Bulasara et al. 2009; Kumana & Associates 2013

Internal Calculations

Catalytic Hydrocracking EIA 2012a; EIA 2013c

Based on downstream charge capacity and 2010 capacity factor of 86.4%

DOE 2007, HP 2011

LBNL 2005a; SEP 2013; DOE 2013; ExxonMobil 2013; McKinsey 2009; Kumana & Associates 2013



Table A2. References for U.S. Throughput Data of the Nine Petroleum Refining Processes Studied and Energy Intensity Data Used to Calculate

the Current Typical, State of the Art, and Thermodynamic Minimum Energy Consumption Bands

Process Throughput/ Production

Reference(s)

Throughput/ Production Value

Description

CT Energy Intensity

Reference(s)

SOA Energy Intensity Reference(s)

TM Energy Intensity Reference

Catalytic Reforming EIA 2012a; EIA 2013c


DOE 2007


DOE 2006

Coking/Visbreaking EIA 2012a; EIA 2013c


DOE 2007

LBNL 2005a; SEP 2013; DOE 2013; ExxonMobil 2013; McKinsey 2009; NRC 2005; Kumana & Associates 2013


Fluid Catalytic Cracking EIA 2012a; EIA 2013c


DOE 2007, HP 2011

LBNL 2005a; SEP 2013; DOE 2013; ExxonMobil 2013; McKinsey 2009; NRC 2004; Al-Mutairi 2010; UOP 2009

DOE 2006

Hydrotreating EIA 2012a; EIA 2013c


DOE 2007



Isomerization EIA 2012b; EIA 2013c

Based on isobutane, isopentane, and isohexane production capacity and 2010 capacity factor of 86.4%

DOE 2007



Vacuum Crude Distillation EIA 2012a; EIA 2013c


DOE 2007

LBNL 2005a; SEP 2013; DOE 2013; ExxonMobil 2013; McKinsey 2009; LBNL 2005b; Kumana & Associates 2013




Appendix A3: Technologies Analyzed to Estimate Practical

Minimum Energy Intensities with References

Table A3. Technologies Analyzed to Estimate Practical Minimum Energy Intensities

Technology Name

Technology Description

Applicabi-lity (Product, process, sector)

Source (See

Reference list at end)

Reported Energy Savings

(Literature-reported

savings, Btu, %, etc.)

Explanation of Savings Baseline, or Reference

(Adjustment, conversion, scale up of reported savings)

Calculated Product/ Process Savings

(Savings compared to SOA or CT energy intensity. PM

savings estimate.)

PM Energy Intensity

(Btu/bbl) or % savings

Atmospheric/Vacuum Crude Distillation

Thermal Cracking

Replacement for crude distillation; alternative to primary separation. Separates crude oil into fractions by cracking large hydrocarbon molecules into smaller ones, thus lowering their boiling points.

Crude distillation

Szklo & Schaeffer 2007, ANL

1999

63,266 Btu/bbl feed (25%)

The replacement of the crude distillation scheme with a thermal cracking process results in a net energy savings of about 63,266 Btu/bbl of oil processed (ANL 1999). Not including electricity losses, hydrogen, and produced energy, the processes would be 199,159 Btu/bbl for crude distillation and 148,407 Btu/bbl for thermal cracking, which would be about a 25% reduction.

Assume a 25% reduction over CT for atmospheric and vacuum distillation. Applying this technology would result in a PM of 81,800 Btu/bbl for atmospheric distillation and 66,800 Btu/bbl for vacuum distillation.

81,800 (atmos-pheric), 66,800

(vacuum)

Progressive Distillation

Integrates atmospheric and vacuum distillation columns; atmospheric distillation, vacuum distillation, gasoline fractionation, and naphtha stabilizer

Only new distillation units (not retrofits)

Szklo & Schaeffer 2007, HP

2008, LBNL 2005

30% savings on total energy use for crude distillation units (Szklo & Schaeffer 2007); 35% reduction in fuel use (LBNL 2005)

Only applicable to distillation units to be constructed or large crude distillation expansion projects

Assume the lower savings of 30% total savings over CT. Applying this technology would result in a PM of 76,400 Btu/bbl for atmospheric distillation and 62,400 Btu/bbl for vacuum distillation.

76,400 (atmos-pheric), 62,400

(vacuum)



Technology Name



Source (See









savings estimate.)

PM Energy Intensity


Atmospheric/Vacuum Crude Distillation (continued)

Self-heat Recuperation

Whole-process heat is recirculated within the process without heat addition, leading to large energy savings

Atmospheric crude distillation

Kansha et al. 2012

48% energy consumption compared to conventional atmospheric distillation

Reduction in energy consumption compared to conventional distillation, requires development of compressors that work at high temperature

Assume a 48% savings when compared to CT for atmospheric distillation (109,100 Btu/bbl). Applying this technology would result in a PM of 56,700 Btu/bbl for atmospheric distillation.

56,700

Dividing-wall columns

Integrates two conventional distillation columns into one by inserting a vertical partition in the central section. The column can contain trays or packing and can handle more than three components.

Crude distillation

Szklo & Schaeffer

2007, IPPC 2012

15% saving potential (total fuel consumption) (Szklo & Scheaffer, 2007); Can reduce energy costs by 30%

Dividing-wall columns can save up to 30% in energy costs and reduce total fuel consumption by 15%

Assume a 25% reduction over CT for atmospheric and vacuum distillation. Applying this technology would result in a PM of 92,700 Btu/bbl for atmospheric distillation and 75,700 Btu/bbl for vacuum distillation.

92,700 (atmos-pheric), 75,700

(vacuum)



Technology Name



Source (See









savings estimate.)

PM Energy Intensity


Hydrotreating

Biodesulfur-ization

Biological removal of sulfur from gasoline that is an alternative to hydrodesulfurization/ hydrotreating. Biodesulfurization utilizes a biological agent (or catalyst) for the removal of sulfur, rather than the conventional cobalt or molybdenum catalysts commonly deployed for hydrodesulfurization. The use of the biological agent results in lower temperatures and atmospheric pressure and eliminates the need for hydrogen as a feed and combustion of fuel gas

Hydro-desulfur-ization/ hydro-treating (certain fuels)


1999, DOE 2003

70-80% (Szklo & Schaeffer 2007), 84% (ANL 1999) decrease in energy use

ANL 1999 estimate based upon using BDS process for desulfurization of about 50% of diesel fuel having a sulfur content greater than .05%. (HDS 356,000 Btu/bbl, BDS 56,000 Btu/bbl)

Comparing ANL 1999's CT for hydrotreating of 56,000 Btu/bbl to the CT of 80,800 Btu/bbl, the savings would be 31%. Applying this technology would result in a PM of 55,800 Btu/bbl.

55,800



Technology Name



Source (See









savings estimate.)

PM Energy Intensity


Utilities - CHP, Boilers, Pumps, etc.

Microturbines Microturbines 26-30% efficient; 40% recovery and can push CHP up to 80% efficiency

w/CHP LBNL 2000

14% increase in efficiency over typical CHP efficiency by adding microturbines

14% increase in efficiency of CHP Systems

Referencing 2006 Petroleum Refining Energy Footprint, 320 TBtu of direct end use is from CHP systems, which equates to 10% of plant wide energy use. 14% savings of 10% energy use results in 1.4% average savings in a typical refinery. Practical minimum specific energy savings of 1.4% over CT applied to all processes.

1.4% savings

over CT for all

processes

Crosscutting Technologies

New High-Temperature, Low-Cost Ceramic Media for Natural Gas Combustion Burners

Combining four different technologies into a single radiant burner package that functions as both a burner and a catalyst support.

Could potentially apply when electric or natural gas radiant heaters used in process heating.

DOE 2011 25% reduction in energy for process heat

Potential to reduce energy consumption by 25% for process heat.

Referencing 2006 Petroleum Refining Energy Footprint, 2,346 TBtu of direct end use for process heating. This equates to 83% of direct end use. 25% savings of 83% energy use results in 21% average savings. Practical minimum specific energy savings of 21% over CT applied to all process units.

21% savings

over CT for all

processes



Technology Name



Source (See









savings estimate.)

PM Energy Intensity


Crosscutting Technologies (continued)

Fouling minimization

Predicting fouling threshold and controlling heat exchanger fouling

Heat exchangers


1999, DOE 1999

2% improvement in overall refinery energy use (ANL 1999, Szklo & Schaeffer 2007)

The 2% improvement applies to all energy use, so assume 2% savings for each process unit. Practical minimum energy savings would be 2% over CT applied to all process units.

2% savings over CT for all process

units

Advanced Energy and Water Recovery Technology from Low-Grade Waste Heat

Technology involves the recovery of high purity water and energy from low grade heat, high moisture waste streams using nanoporous membranes. Concept will be proven in laboratory and evaluates in "two different types of industrial environments.”

Applies to refineries that utilize wet scrubbers

DOE 2011; GTI

2011

Estimated 20-30% greater energy efficiency in recovery from low grade waste heat; 18.9 TBtu/year for the refining industry

An energy efficiency gain of 20-30% could be achieved when a transport membrane condenser is used along with water recovery

There will be an estimated 18.9 TBtu/year energy savings for the refining industry with wet scrubbers. Compared to the overall CT energy consumption of 3,176 TBtu/year, this represents a 1% energy savings. Practical minimum energy savings of 1% over CT is applied to all process units.

1% savings over CT for all process

units



Technology Name



Source (See









savings estimate.)

PM Energy Intensity



Higher efficiency motors and lubricants, higher efficiency burners, and better heat integration

1

All process units

UOP 2013

Range of savings for each process, accounting for 3% electricity, 10% fuel gas, and 20% steam reductions

These technology improvements would enable the following reductions: 3% reduction in electricity (higher efficiency motors and lubricants), 10% reduction in fuel gas (higher efficiency burners), and 20% reduction in steam (better heat integration)

Overall, the total energy savings for all process units would be 12%. Practical minimum specific energy savings of 12% over CT overall, with savings for the individual process units ranging from 2-18% over CT.

2-18% for individual process

units1, 12%

overall


1 Savings for individual process units for this technology improvement are as follows: 12% for alkylation, 14% for atmospheric crude distillation, 17% for coking/visbreaking, 2% for fluid catalytic

cracking, 7% for catalytic hydrocracking, 10% for hydrotreating, 18% for isomerization, 7% for catalytic reforming, and 15% for vacuum crude distillation.


REFERENCES FOR TABLE A3

ANL 1999

The Potential for Reducing Energy Utilization in the Refining Industry.

Argonne National Laboratory. Petrick, M. and Pellegrino, J. 1999.

www.ipd.anl.gov/anlpubs/1999/10/34061.pdf

DOE 1999

“Effective Fouling Minimization Increases the Efficiency and Productivity of

Refineries.” U.S. Department of Energy, Office of Industrial Technologies.

January 1999. www.energy.gov/sites/prod/files/2014/05/f16/foulmitpet.pdf

DOE 2003

“Gasoline Biodesulfurization.” U.S. Department of Energy, Industrial

Technologies Program. May 2003.

http://energy.gov/sites/prod/files/2014/05/f16/gasbiopet_final.pdf

DOE 2011

Grand Challenge Portfolio: Driving Innovations in Industrial Energy

Efficiency. U.S. Department of Energy, Industrial Technologies Program.

2011.

www1.eere.energy.gov/manufacturing/pdfs/grand_challenges_portfolio.pdf

GTI 2011

Final Technical Report: Advanced Energy and Water Recovery Technology

from Low Grade Waste Heat. Gas Technology Institute. 2011.

www.osti.gov/bridge/servlets/purl/1031483/1031483.pdf

HP 2008 “Improved crude oil fractionation by distributed distillation”, Hydrocarbon

Processing (HP), Haddad, H.N. and Manley, D.B. 2008.

IPPC 2012

Best Available Techniques (BAT) Reference Document for the Refining of

mineral oil and gas. Final Draft. European Commission. Integrated Pollution

Prevention and Control (IPPC) Bureau. 2013.

http://eippcb.jrc.ec.europa.eu/reference/BREF/FD_REF_July_2013online.pdf

Kansha et al.

2012

“Application of the self-heat recuperation technology to crude oil

distillation.” Kansha, Y., Kishimoto, A., and Tsutsumi, A. Applied Thermal

Engineering 43(2012): 153-157. 2012.

LBNL 2000

Emerging Energy-Efficient Industrial Technologies. Lawrence Berkeley

National Laboratory (LBNL). Martin, N., Worrell, E., Ruth, M., Price, L.,

Elliott, R.N., Shipley, A.M., Thorne, J. 2000.

https://escholarship.org/uc/item/5jr2m969#page-2

LBNL 2005

Energy Efficiency Improvement and Cost Saving Opportunities for Petroleum

Refineries. Worrell, E. and Galitsky, C. Lawrence Berkeley National

Laboratory (LBNL). 2005.

Szklo &

Schaeffer 2007

“Fuel specification, energy consumption, and CO2 emissions in oil

refineries.” Energy 32: 1075-1092. Szklo, A. and Schaeffer, R. 2007.

UOP 2013 Personal communications. 2013.


Appendix A4: Practical Minimum Technology

Weighting Factors

METHODOLOGY TO DETERMINE WEIGHTING FACTORS

In this section the practical minimum technology weighting factors methodology is explained.

The application of this methodology is presented in Table A4.

Six Weighting Factors, A through F, are considered for each technology and scored as shown

(High (H) = 3, Medium (M) = 2, Low (L) = 1, Not Available (A) = 0). The factors are also

scaled according to DOE Importance Level, e.g., an importance level of 2 carries twice the

weight of an importance level of 1. For the petroleum refining bandwidth, factors A-F each

carried a DOE Importance Level of 1.

The DOE Importance Level is multiplied by the score for each factor and divided by the total

possible score to determine overall weighting of technology. The NA score of 0 is excluded from

overall weighting.

Factor A - Technology Readiness

High = Technology Readiness Level (TRL) 7-9

Medium = TRL 4-6

Low = TRL 1-3

Factor B - Market Impact

High = widely applicable to all establishments

Medium = applicable to many establishments

Low = applicable to select few establishments or unique process

Factor C - Relative Cost and Savings Payback

High = implementation cost >90% of reference technology, or payback > 10 years

Medium = cost <90% and >40% of reference technology, payback <10 years

Low = cost <40% of reference, payback < 2 years

Note: the score is reversed such that H = 1 and L = 3

Factor D – Technical Risk

High = high likelihood of technology success and deployment, minimal risk factors

Medium = insufficient evidence of technology success, some risk factors

Low = low likelihood of success, multiple and significant risk factors

Note: the score is reversed such that H = 1 and L = 3


Factor E – Productivity/Product Quality Gain

High = significant gain in productivity, either quantity or quality of product produced

Medium = moderate gain in productivity

Low = no gain in productivity

Factor F – Environmental Benefits

High = multiple and significant environmental benefits,

Medium = some environmental benefits,

Low = little or no environmental benefit


Table A4. Practical Minimum Technologies Analysis with Weighting Factors

Importance Level

1 1 1 1 1 1

Technology Name

Technology Weighting Factors

Overall Importance

Rating

A – Technology Readiness

B- Market Impact C- Relative Cost

and Savings Payback

D- Technical Risk E – Productivity/ Product Quality

Gain

F- Environmental Benefits

H, M, or L

Explanation H, M, or L





Explanation

Atmospheric and Vacuum Crude Distillation

Thermal cracking

L Engineering

judgment M

Applicable to

distillation, but would

involve replacement

H

Noted as 'high-cost' to replace all

towers

H Complete

replacement of distillation

L Insufficient information

M

Could remove

contaminants (e.g.,

sulfur) and reduce CO2 emissions

44%

Dividing-wall columns

H

Are some commercial

units, 'proven

technology'

H Widely

applicable M

Lower capital costs

M

Technology still needs

more development


L Engineering

judgment 67%

Progressive Distillation

H

Applied to 2 refineries

(LBNL 2005) -TRL 8

M

Applicable to


involve replacement

M Lower cost H

Applicable to new

refineries only


L Engineering

judgment 56%

Self-heat recuperation

L Engineering judgment –

TRL 1 M

Applicable to


involve replacement

NA Insufficient information

H

More optimization

and modification

needed



40%



Importance Level

1 1 1 1 1 1

Technology Name


Overall Importance

Rating



and Savings Payback


Gain


H, M, or L






Explanation

Hydrotreating

Biodesulfur-ization

L Engineering Judgment -

TRL 1 L

Applicable depending on sulfur content

M

Lower capital and operating

costs

L

Would require

significant breakthroug

h


M

Improved environment

al performance

56%

Utilities - CHP, Boilers, etc.

Microturbines H Engineering Judgment -

TRL 9 L

Small targeted

application H

Major capital investment

M Moderate process change

NA Engineering

judgment H

Large energy savings

56%

Crosscutting Technologies

New High-Temperature, Low-Cost Ceramic Media for Natural Gas Combustion Burners

H Engineering Judgment -

TRL 7 H

Wide ranging

applications M

Moderate capital

investment M

Moderate process change

M Better

heating H

Large energy savings

83%



Importance Level

1 1 1 1 1 1

Technology Name


Overall Importance

Rating



and Savings Payback


Gain


H, M, or L






Explanation


Fouling Minimization

M Engineering Judgment

H Applicable

to all refineries

NA Insufficient information

L

Small, well understood

process change

M Engineering

judgment L

Engineering judgment

73%

Advanced Energy and Water Recovery Technology from Low-Grade Waste Heat

M Engineering Judgment -

TRL 4 H

Wide ranging

applications H

Major capital investment

H Large

process change

L Engineering

judgment H

Waste water recovery

61%

Appendix A4 provides the methodology used to identify the weighting factors and the definitions for the abbreviations.


Appendix A5: Thermodynamic Minimum

Calculation Details

This Appendix provides details on how the thermodynamic minimum energy intensities for

atmospheric and vacuum crude distillation, hydrotreating, catalytic hydrocracking, isomerization,

and coking/visbreaking were calculated and assumptions and reference values used. The

thermodynamic minimum energy intensities for catalytic reforming, alkylation, and fluid

catalytic cracking were taken from the previous bandwidth, DOE 2006.

ATMOSPHERIC AND VACUUM CRUDE DISTILLATION THERMODYNAMIC

MINIMUM ENERGY INTENSITY

Distillation takes advantage of differences in boiling points to separate crude oil. The overall

heat balance is described by:

Thermodynamic Minimum Energy Intensity = Energy Out – Energy In

It is assumed for these TM energy intensity calculations that:

Crude oil behaves as an “ideal solution”; that is, the properties of the component in solution

are equal to the sum properties of the pure component

The heavier fractions must be distilled under vacuum (10 millimeters mercury (mm Hg)) to

prevent the heavy fractions from degrading

No work is spent establishing or maintaining the vacuum

Crude fractions enter at room temperature and exit at their respective boiling points, except

for residual oil which exits at the temperature of the boiler

The stream enters and exits as a liquid

The process is adiabatic meaning that no heat is lost to the environment

All processes were considered to be completely reversible

Heat recovery is limited by the processes’ Carnot efficiency.

Change in entropy (ΔS) was calculated at 842°F (450 °C)

Heat capacities are those for ideal gases

During the distillation process, the crude oil is heated so that the lighter fractions evaporate,

which allows the vapor to rise up through the column until it contacts a tray that is at the vapor

component’s boiling point. The component condenses and exits the column as a liquid stream.

Therefore, the energy input is the amount of energy required to raise the temperature of each

component from the ambient temperature (e.g., 77oF / 25

oC) to its boiling point. The energy

required to evaporate the crude oil component is cancelled out by the energy released when the

component vapor condenses. As an ideal solution, the boiling point of the pure substance is used

and any effects of intermolecular interactions are ignored.


The TM energy intensity was calculated as the change in Gibbs free energy, ΔGdist (heat of

reaction and Gibbs free energy are related as follows: ΔG = ΔH –TΔS, where T is absolute

temperature (either Kelvin or Rankine), ΔH is the change in enthalpy, and ΔS is the change in

entropy).

The energy consumed by atmospheric distillation includes energy that goes into heating the

heavy fractions that must be distilled under vacuum. However, for the TM energy intensity

calculation, the energy consumption of atmospheric distillation is limited to the separation

energy for the crude fractions that can be distilled at atmospheric pressure, and the work done to

establish or maintain a vacuum. In addition, the calculation excludes the energy content of the

fuel gas stream generated via atmospheric distillation and excludes the heat recovery that takes

place via the crude preheat train.

The vacuum distillation process is also simplified to calculate the TM energy intensity. Similar

to atmospheric distillation, it is assumed that all energy consumed by the vacuum distilled

fractions as they are heated from ambient temperature to their boiling points is included in the

vacuum distillation TM energy intensity. In reality, the heavy components are heated from

ambient conditions to a higher temperature as they pass through the atmospheric distillation

tower. In addition, it is assumed that the residue stream produced is processed further in coking

units, rather than used to generate heat for the vacuum distillation tower. Table A5 shows the

physical and chemical properties of the crude oil fractions.


Table A5. Sample Product Fractions from Barrel of Crude Oil Feedstock, Used in Calculating Distillation

Thermodynamic Minimum Energy Intensity

Product

Temperature

Cut

Volume

% of

Crude

Specific

Gravity

Volume

Average

boiling

point

(°F)

Watson

Factor

(Kw)

Cp

(Btu/

°F*lb)

ΔHheating

(Btu/bbl)

Atm

osp

he

ric

Dis

tilla

tio

n

LPG <=C4 1.57 0.57 37 NA 0.19 0

Light Straight

Run

C5-165 °F

(C5-74 °C) 8.26 0.67 98 12.3 0.54 245

Heavy Straight

Run

165-331 °F

(74-166 °C) 20.96 0.76 248 11.7 0.54 5,015

Light Middle

Distillate

331-896 °F

(166-480 °C) 17.11 0.81 479 12 0.59 11,454

Heavy Middle

Distillate

896-480 °F

(480-249 °C) 17.52 0.85 529 11.8 0.58 14,193

Total 65.42% 30,907

Va

cu

um

Dis

tilla

tio

n

Vacuum Gas

Oil

480-999 °F

(249-537 °C) 24.71 0.90 791 12.0 0.64 13,526

Residual ≥999 °F

(≥537 °C) 9.87 0.99 999 11.7 0.64 20,468

Total

34.58% 33,394

Distillation Total

100% 0.83 490 64,897

Sources: Riazi 2007; Chang et al. 2012; Nelson 1958

Notes: “C” refers to number of carbon atoms in the hydrocarbon; rounding errors present.

The energy required to raise the temperature of each fraction to its boiling point (bp) is

calculated by:

ΔHfraction = mCpΔT = masscrude fraction * Cp * (Tbp – 77°F)

Where m is mass, Cp is specific heat capacity, and Tbp is the boiling point temperature. The ΔS is

calculated by the following equation:

Where xi is the mole fraction of a given species.

This equation yields a value of ΔS=-9.85 Btu/K for the sample barrel depicted in Table A5. This

value is negative because there is a decrease in entropy (disorder) in the system. The separation

of a mixture, even an ideal one, typically requires an input of energy.

Thus the ΔGdist is calculated as follows

ΔGdist= ΔH – TΔS = 64,897 – 842*(-9.85) = 69,312 Btu/bbl of crude oil


This value is the least possible energy that must be spent to heat all of the fractions to their

boiling point and to separate the incoming feed stream to its components. However, there are

energy losses which can be systemically accounted for because of Carnot inefficiencies. The

energy used to boil the component is not completely recovered upon condensation in a

distillation process. Only a fraction of the ΔHvap can be recovered according to the equation:

where qinput= ΔHvap, Tb is the boiler temperature, and Tc is the condenser temperature on an

absolute scale.

Assuming a boiler operating at 660°F (atmospheric) or 1063°F (vacuum) and the condenser

operating at the average boiling point of each fraction gives a total energy loss of 1.4 Btu/bbl for

the distillation process. Accounting for this known limit to distillation leads to theoretical process

minimum, which accounts for the previously calculated TM energy intensity and the limits of the

distillation process resulting in ΔGdist=69,313 Btu/bbl.

HYDROTREATING THERMODYNAMIC MINIMUM ENERGY INTENSITY

Catalytic hydrotreating is the process by which hydrogen is added to reduce double bonds and to

remove sulfur, nitrogen, oxygen and halogen impurities from the feedstock. It is assumed for the

TM calculation that:

The entire hydrotreater process occurs at 570 °F

300 standard cubic feet (scf) of H2 is consumed per barrel

All streams consist of ideal gases

Sample barrel characteristics:

o Sulfur: 0.3675 weight percent (wt%)

o Mercaptan 25 parts per million (ppm)

o Nitrogen 970 ppm

o No halogens or oxygen

o 75% of remainder H2 goes to reduce molecules with 1 double bond

o 25% of remainder H2 goes to reduce molecules with 2 double bonds

Hydrotreaters are run between 300 and 400 °C (assumed 570 °F for entropy calculations) which

is insufficient to permit cracking.

To calculate the change in energy, ΔGrxn was calculated for each reaction at the temperature of

the reactor. Temperature only impacts entropy which was assumed to be the change in moles of

ideal gas. No change in the temperature of the feed stream was assumed through the length of the

reactor. For the sample barrel: sulfur was 0.3675 wt% (evenly divided between sulfide, disulfide

and thiophene), nitrogen was 970 ppm and mercaptan sulfur was 25 ppm. The balance of the

hydrogen consumed was assumed to go to saturating olefins.


Table A6. Change in Gibbs Free Energy for Hydrotreating Reactions

Reaction ΔGrxn

(Btu/scf) Cubic Feet

of H2a

ΔGrxn

(Btu/bbl)

Hydrogenation of one double bond -180.64 215.06 -38,849

Hydrogenation of two double bonds -181.46 71.70 -13,007

Desulfurization of sulfide -64.77 1.81 -117

Desulfurization of mercaptan -65.90 0.02 -2

Desulfurization of disulfide -56.72 2.71 -154

Desulfurization of thiophene -123.94 3.62 -448

Denitrogenation of pyrrole -142.51 2.30 -328

Denitrogenation of pyridine -32.08 2.88 -92

Deoxidation of phenol -133.01 0 0

Deoxidation of peroxides -383.30 0 0

Dehalognenation (chlorine) -100.65 0 0

Hydrocracking -47.93 0 0

Total 300 -53,000

a Volume at standard temperature and pressure (25°C and 1 atmospheric pressure)

Source: Chang et al. 2012 and ChemEd DL n.d.

The energy consumed by a hydrotreater is strongly related to the degree of saturation of the feed

stream and the amount of impurities in a barrel of oil. Given a set ratio of impurities, the majority of

H2 is consumed in reducing C-C double bonds. For these calculations 75% of unsaturated molecules

were assumed to have one double bond and 25% to have two double bonds. The small difference

between the two free energies of reaction is the result of entropy differences in the two reactions.

With 300 scf of H2 consumed in the feed stream, the ΔG or final TM energy intensity value is equal

to -53,000 Btu/bbl. A typical hydrocracker consumes between 200 and 800 scf H2/bbl. This

corresponds to the range ΔG between -35,000 and -143,000 Btu/bbl. Thus the degree of unsaturation

plays a large role in determining the TM of a hydrocracker. While all reactions at this temperature are

exoergonic, there is nearly six-fold greater energy released from the hydrogenation of a double bond

as opposed to the denitrogenation of pyridine. With greater saturated feedstocks the concentration of

impurities plays a larger role in determining the TM energy intensity.

CATALYTIC HYDROCRACKING THERMODYNAMIC MINIMUM ENERGY

INTENSITY

Hydrocracking is an exothermic process, which replaces a carbon-carbon single bond and a

hydrogen-hydrogen single bond with two carbon-hydrogen bonds. This reduces the chain lengths

of hydrocarbons to create more valuable molecules. It is assumed for the hydrocracking TM

energy intensity calculation that:

The entire hydrocracking process occurs at 572°F (300 °C)


1,500 scf of H2 is consumed per barrel

All streams consist of ideal gases

No processes other than cracking occur in the hydrocracker

Entropy is dominated by the change in moles of gas

As in Table A6, for each cubic foot of H2 consumed -47.93 Btu of energy are released in the

hydrocracking of a molecule. The sample barrel used (see Table A5) consumes 1,500 scf of H2.

Thus the TM for hydrocracking is equal to -71,900 Btu/bbl, although this number varies directly

with the amount of hydrogen needed to hydrocrack the incoming feed stream.

ISOMERIZATION THERMODYNAMIC MINIMUM ENERGY INTENSITY

In the model for the isomerization process, the following incoming and outgoing streams were

assumed as follows in Table A7:

Table A7. Isomerization Process Stream Assumptions

Chemical Input

(wt%)

Output

(wt%)

ΔGf

(Btu/mol)

ΔG

(Btu/bbl)

Isopentane 22 41 -146 -38.66

n-Pentane 33 12 -139 40.86

2,2-Dimethylbutane 1 15 -176 -28.86

2,3-Dimethylbutane 2 5 -169 -5.92

2-Methylpentane 12 15 -166 -5.81

3-Methylpentane 10 7 -163 5.72

n-Hexane 20 5 -158 27.76

Total 100 100 -4.91

Source: Gary et al. 2007


It is also assumed for this TM energy intensity calculation that:

The change in volume is negligible

No dehydrogenation or cracking occurs

Only five and six carbon molecules rearrange

The feed and exit stream is fully characterized by Table A7

Streams enter and exit the process at the same temperature

No change in entropy since the number of molecules remains the same

The thermodynamic minimum is the change in free energy that occurs between the input and

output streams. Branched alkanes are energetically more stable, thus this process is exoergonic.

For a barrel consisting of these inputs and outputs the ΔG is -4.91 Btu/bbl. This number is

strongly dependent upon which branched alkanes are favored by the catalyst and operating

conditions.

COKING/VISBREAKING (DELAYED COKING) THERMODYNAMIC

MINIMUM ENERGY INTENSITY

Because delayed coking consumes a majority of the energy for the coking/visbreaking process

unit, the TM energy intensity for delayed coking was assumed to also be the TM energy intensity

for all coking/visbreaking. The calculation of the TM for delayed coking is described in this

section. For coking, the change in entropy is proportional to the natural log of the ratio of the

number of moles after and before coking. With a temperature of 600 °F, the ΔS= -3.7 Btu/bbl.

Combined with the ΔH from the sample barrel, the ΔG or TM for the delayed coking process is

345.0 Btu/bbl. Coking is an endoergonic process meaning it requires energy to proceed. A list of

assumptions and calculations can be found in Table A8 and Table A9 below.

It is also assumed for this TM energy intensity calculation that:

Entropy is dominated by the change in moles; for every molecule that enters, about 2.1 exit

Coking proceeds at 600 °F

No chlorine, nitrogen, oxygen or metals are in the feed stream

15% unsaturation in gasoline and 20% unsaturation in gas oil for the exit stream

No saturation in residual crude feed or coke stream

LPG unsaturation is as noted in Table A5

Sample delayed coker input/output is based on crude oil from North Slope, Alaska in Gary et

al. 2007

Production volumes are based upon 100,000 BPCD

Conradson Carbon = 19%


Table A8. Coking Process Assumptions and Calculations

Component Stream

Make-up

(vol%)

Production Rate

(lb/hour)

Sulfur Content

(wt%)

C/H ratio

Avg. C length

Degree of Unsaturation

Total ΔH in Bonds

(Btu/bbl)

Inp

ut 1000°F+

Residual Crude

100% 345,080 2.30 1.3 30.0 15% -9,919

Ou

tpu

t

Gas 10.5 wt% 36,230 6.50 see Table A9 below 771

Gasoline 23.3% 61,420 0.65 2.3 8.0 15% 2,373

Gas Oil 45.2% 142,530 1.95 1.8 18.0 20% 4,889

Coke 30.4 wt% 104,900 2.27 0.5 50.0 30% 2,234

Total 100.0% 345,080 349

ΔG = 345.0 Btu/bbl

Source: Gary et al. 2007, Colorado School of Mines n.d., AIChE 2000; Bond enthalpies are from ChemEd DL n.d.

Table A9. Calculations Used to Determine Coking

Entropy

Component Production

Rate

(lb/hour)

Total ΔH in

Bonds

(Btu/bbl)

Methane 12,580 270.4

Ethylene 640 13.8

Ethane 7,300 163.2

Propylene 1,990 44.1

Propane 5,520 125.1

Butylene 2,060 46.2

i-butane 890 20.3

n-butane 2,310 52.7

H2 270 25.6

CO2 140 1.5

H2S 2,530 8.3

Total 36,230 771.2


THERMODYNAMIC MINIMUM ENERGY INTENSITY REFERENCES

AIChE 2000 Refining Overview- Petroleum Process & Products (CD-ROM). Self, F., Ekholm,

E., and Bowers, K. American Institute of Chemical Engineers (AIChE). 2000.

Chang et al.

2012

“Characterization, Physical and Thermodynamic Properties of Oil Fractions.” In

Refinery Engineering: Integrated Process Modeling and Optimization, 1st ed.

Chang, A.F., Pashikanti, K., and Y.A. Liu. 2012. Wiley-VCH Verlag GmbH & Co.

www.wiley-vch.de/books/sample/3527333576_c01.pdf

ChemEd DL

n.d. “Bond Enthalpies.” Chemical Education Digital Library (ChemEd DL). n.d.

Colorado

School of

Mines n.d.

“Delayed Coking: Chapter 5.” Presentation. Colorado School of Mines. n.d.

DOE 2006

Energy Bandwidth for Petroleum Refining Processes. Prepared by Energetics, Inc.

for U.S. Department of Energy (DOE), Office of Industrial Technologies (OIT).

2006. http://energy.gov/sites/prod/files/2013/11/f4/bandwidth.pdf

Gary et al.

2007

Petroleum Refining: Technology and Economics, 5th

edition. Gary, J.H., Handwerk,

G.E., and Kaiser, M.J. New York, NY: Marcel Dekker, Inc. 2007.

Nelson 1958 Petroleum Refinery Engineering, 4

th ed. Nelson, W.L. New York: McGraw Hill.

1958.

Riazi 2007 Characterization and Properties of Petroleum Fractions. Riazi, M.R. ASTM Stock

Number: MNL 50. USA: ASTM International. 2007.

bandwidth study on energy use and potential energy … study on energy use and potential energy...

Documents